repo
stringlengths
2
99
file
stringlengths
13
225
code
stringlengths
0
18.3M
file_length
int64
0
18.3M
avg_line_length
float64
0
1.36M
max_line_length
int64
0
4.26M
extension_type
stringclasses
1 value
mff
mff-master/mff/kernels/base.py
import os from abc import ABCMeta, abstractmethod from pathlib import Path path = Path(os.path.abspath(__file__)) Mffpath = path.parent.parent / "cache/" class Kernel(metaclass=ABCMeta): @abstractmethod def __init__(self, kernel_name, *args, **kwargs): super().__init__(*args, **kwargs) self.kernel_name = kernel_name
345
23.714286
53
py
mff
mff-master/mff/kernels/twobodykernel.py
# -*- coding: utf-8 -*- import logging import os.path import pickle from abc import ABCMeta, abstractmethod import numpy as np from mff.kernels.base import Kernel, Mffpath logger = logging.getLogger(__name__) def dummy_calc_ff(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype = data if kertype == "single": with open(Mffpath / "k2_ff_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "k2_ff_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros((len(array), 3, 3)) for i in np.arange(len(array)): result[i] = fun(np.zeros(3), np.zeros(3), array[i][0], array[i][1], theta0, theta1, theta2) return result def dummy_calc_ee(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype, mapping = data if kertype == "single": with open(Mffpath / "k2_ee_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "k2_ee_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros(len(array)) if not mapping: for i in np.arange(len(array)): for conf1 in array[i][0]: for conf2 in array[i][1]: result[i] += 0.25*fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) else: for i in np.arange(len(array)): for conf2 in array[i][1]: result[i] += 0.5*fun(np.zeros(3), np.zeros(3), array[i][0], conf2, theta0, theta1, theta2) return result def dummy_calc_ef(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype, mapping = data if kertype == "single": with open(Mffpath / "k2_ef_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "k2_ef_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros((len(array), 3)) if not mapping: for i in np.arange(len(array)): conf2 = np.array(array[i][1], dtype='float') for conf1 in array[i][0]: conf1 = np.array(conf1, dtype='float') result[i] += -0.5*fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) else: for i in np.arange(len(array)): conf2 = np.array(array[i][1], dtype='float') conf1 = np.array(array[i][0], dtype='float') result[i] += -fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) return result class BaseTwoBody(Kernel, metaclass=ABCMeta): """ Two body kernel class Handles the functions common to the single-species and multi-species two-body kernels. Args: kernel_name (str): To choose between single- and two-species kernel theta[0] (float) : lengthscale of the kernel theta[1] (float) : decay rate of the cutoff function theta[2] (float) : cutoff radius bounds (list) : bounds of the kernel function. Attributes: k2_ee (object): Energy-energy kernel function k2_ef (object): Energy-force kernel function k2_ff (object): Force-force kernel function """ @abstractmethod def __init__(self, kernel_name, theta, bounds): super().__init__(kernel_name) self.theta = theta self.bounds = bounds self.k2_ee, self.k2_ef, self.k2_ff = self.compile_theano() def calc(self, X1, X2, ncores=1): """ Calculate the force-force kernel between two sets of configurations. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N2*3 matrix of the vector-valued kernels """ ker = np.zeros((len(X1) * 3, len(X2) * 3)) if ncores > 1: confs = [] for x1 in X1: for x2 in X2: confs.append(np.asarray([x1, x2])) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 2-body force-force kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ff, clist) pool.close() pool.join() result = np.concatenate(result).reshape((n, 3, 3)) for i in range(len(X1)): for j in range(len(X2)): ker[i * 3: i * 3 + 3, 3 * j:3 * j + 3] = result[(j + i * len(X2))] else: for i, conf1 in enumerate(X1): for j, conf2 in enumerate(X2): ker[i * 3:i * 3 + 3, 3 * j:3 * j + 3] += self.k2_ff( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_ef(self, X_glob, X, ncores=1, mapping=False): """ Calculate the energy-force kernel between two sets of configurations. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N2*3 matrix of the vector-valued kernels """ ker = np.zeros((len(X_glob), len(X) * 3)) if ncores > 1: confs = [] for x1 in X_glob: for x2 in X: confs.append(np.asarray([x1, x2])) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 2-body energy-force kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, mapping] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ef, clist) pool.close() pool.join() result = np.vstack(np.asarray(result)) for i in range(len(X_glob)): for j in range(len(X)): ker[i, 3 * j:3 * j + 3] = result[(j + i * len(X))] else: if not mapping: for i, x1 in enumerate(X_glob): for j, conf2 in enumerate(X): for conf1 in x1: ker[i, 3 * j:3 * j + 3] += 0.5*self.k2_ef( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) else: for i, conf1 in enumerate(X_glob): for j, conf2 in enumerate(X): ker[i, 3 * j:3 * j + 3] += self.k2_ef( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_ee(self, X1, X2, ncores=1, mapping=False): """ Calculate the energy-energy kernel between two global environments. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N1 x N2 matrix of the scalar-valued kernels """ if ncores > 1: # Used for multiprocessing confs = [] # Build a list of all input pairs which matrix needs to be computed for x1 in X1: for x2 in X2: confs.append(np.asarray([x1, x2])) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 2-body energy-energy kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, mapping] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ee, clist) pool.close() pool.join() result = np.concatenate(result).ravel() ker = np.zeros((len(X1), len(X2))) for i in range(len(X1)): for j in range(len(X2)): ker[i, j] = result[j + i*len(X2)] else: if not mapping: ker = np.zeros((len(X1), len(X2))) for i, x1 in enumerate(X1): for j, x2 in enumerate(X2): for conf1 in x1: for conf2 in x2: ker[i, j] += 0.25*self.k2_ee(conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) else: ker = np.zeros((len(X1), len(X2))) for i, conf1 in enumerate(X1): for j, x2 in enumerate(X2): for conf2 in x2: ker[i, j] += 0.5*self.k2_ee(conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_gram(self, X, ncores=1, eval_gradient=False): """ Calculate the force-force gram matrix for a set of configurations X. Args: X (list): list of N Mx5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N*3 x N*3 gram matrix of the matrix-valued kernels """ if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: # Used for multiprocessing confs = [] # Build a list of all input pairs which matrix needs to be computed for i in np.arange(len(X)): for j in np.arange(i + 1): thislist = np.asarray([X[i], X[j]]) confs.append(thislist) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 2-body force-force gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ff, clist) pool.close() pool.join() result = np.concatenate(result).reshape((n, 3, 3)) off_diag = np.zeros((len(X) * 3, len(X) * 3)) diag = np.zeros((len(X) * 3, len(X) * 3)) for i in np.arange(len(X)): diag[3 * i:3 * i + 3, 3 * i:3 * i + 3] = result[i + i * (i + 1) // 2] for j in np.arange(i): off_diag[3 * i:3 * i + 3, 3 * j:3 * j + 3] = result[j + i * (i + 1) // 2] else: diag = np.zeros((X.shape[0] * 3, X.shape[0] * 3)) off_diag = np.zeros((X.shape[0] * 3, X.shape[0] * 3)) for i in np.arange(X.shape[0]): diag[3 * i:3 * i + 3, 3 * i:3 * i + 3] = \ self.k2_ff(X[i], X[i], self.theta[0], self.theta[1], self.theta[2]) for j in np.arange(i): off_diag[3 * i:3 * i + 3, 3 * j:3 * j + 3] = \ self.k2_ff(X[i], X[j], self.theta[0], self.theta[1], self.theta[2]) gram = diag + off_diag + off_diag.T # The gram matrix is symmetric return gram def calc_gram_e(self, X, ncores=1, eval_gradient=False): """ Calculate the energy-energy gram matrix for a set of configurations X. Args: X (list): list of N Mx5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N x N gram matrix of the scalar-valued kernels """ if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: # Used for multiprocessing confs = [] # Build a list of all input pairs which matrix needs to be computed for i in np.arange(len(X)): for j in np.arange(i + 1): thislist = np.array([list(X[i]), list(X[j])]) confs.append(thislist) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 2-body energy-energy gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, False] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ee, clist) pool.close() pool.join() result = np.concatenate(result).ravel() off_diag = np.zeros((len(X), len(X))) diag = np.zeros((len(X), len(X))) for i in np.arange(len(X)): diag[i, i] = result[i + i * (i + 1) // 2] for j in np.arange(i): off_diag[i, j] = result[j + i * (i + 1) // 2] else: diag = np.zeros((X.shape[0], X.shape[0])) off_diag = np.zeros((X.shape[0], X.shape[0])) for i in np.arange(X.shape[0]): for k, conf1 in enumerate(X[i]): diag[i, i] += 0.25*self.k2_ee(conf1, conf1, self.theta[0], self.theta[1], self.theta[2]) for conf2 in X[i][:k]: # *2 here to speed up the loop diag[i, i] += 0.25*2.0*self.k2_ee( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) for j in np.arange(i): for conf1 in X[i]: for conf2 in X[j]: off_diag[i, j] += 0.25*self.k2_ee( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) gram = diag + off_diag + off_diag.T # Gram matrix is symmetric return gram def calc_gram_ef(self, X, X_glob, ncores=1, eval_gradient=False): """ Calculate the energy-force gram matrix for a set of configurations X. This returns a non-symmetric matrix which is equal to the transpose of the force-energy gram matrix. Args: X (list): list of N1 M1x5 arrays containing xyz coordinates and atomic species X_glob (list): list of N2 M2x5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N2 x N1*3 gram matrix of the vector-valued kernels """ gram = np.zeros((X_glob.shape[0], X.shape[0] * 3)) if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: # Multiprocessing confs = [] for i in np.arange(len(X_glob)): for j in np.arange(len(X)): thislist = np.asarray([X_glob[i], X[j]]) confs.append(thislist) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 2-body energy-force gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, False] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ef, clist) pool.close() pool.join() result = np.vstack(np.asarray(result)) for i in np.arange(X_glob.shape[0]): for j in np.arange(X.shape[0]): gram[i, 3 * j:3 * j + 3] = result[(j + i * X.shape[0])] else: for i in np.arange(X_glob.shape[0]): for j in np.arange(X.shape[0]): for k in X_glob[i]: gram[i, 3 * j:3 * j + 3] += 0.5*self.k2_ef( k, X[j], self.theta[0], self.theta[1], self.theta[2]) self.gram_ef = gram return gram @staticmethod @abstractmethod def compile_theano(): return None, None, None class TwoBodySingleSpeciesKernel(BaseTwoBody): """Two body single species kernel. Args: theta[0] (float): lengthscale of the kernel theta[1] (float): decay rate of the cutoff function theta[2] (float): cutoff radius """ def __init__(self, theta=(1., 1., 1.), bounds=((1e-2, 1e2), (1e-2, 1e2), (1e-2, 1e2))): super().__init__(kernel_name='TwoBodySingleSpecies', theta=theta, bounds=bounds) self.type = "single" @staticmethod def compile_theano(): """ This function generates theano compiled kernels for global energy and force learning The position of the atoms relative to the central one, and their chemical species are defined by a matrix of dimension Mx5 here called r1 and r2. Returns: k2_ee (func): energy-energy kernel k2_ef (func): energy-force kernel k2_ff (func): force-force kernel """ if not (os.path.exists(Mffpath / 'k2_ee_s.pickle') and os.path.exists(Mffpath / 'k2_ef_s.pickle') and os.path.exists(Mffpath / 'k2_ff_s.pickle')): print("Building Kernels") import theano.tensor as T from theano import function, scan logger.info( "Started compilation of theano two body single species kernels") # -------------------------------------------------- # INITIAL DEFINITIONS # -------------------------------------------------- # positions of central atoms r1, r2 = T.dvectors('r1d', 'r2d') # positions of neighbours rho1, rho2 = T.dmatrices('rho1', 'rho2') # lengthscale hyperparameter sig = T.dscalar('sig') # cutoff hyperparameters theta = T.dscalar('theta') rc = T.dscalar('rc') # positions of neighbours without chemical species (3D space assumed) rho1s = rho1[:, 0:3] rho2s = rho2[:, 0:3] # distances of atoms wrt to the central one and wrt each other in 1 and 2 r1j = T.sqrt(T.sum((rho1s[:, :] - r1[None, :]) ** 2, axis=1)) r2m = T.sqrt(T.sum((rho2s[:, :] - r2[None, :]) ** 2, axis=1)) # squared exponential of the above distance matrices se_jm = T.exp(-(r1j[:, None] - r2m[None, :]) ** 2 / (2 * sig ** 2)) cut_jm = 0.5*(1+T.cos(np.pi*r1j[:, None]/rc))*0.5*(1+T.cos(np.pi*r2m[None, :]/rc))*( (T.sgn(rc-r1j[:, None]) + 1) / 2)*((T.sgn(rc-r2m[None, :]) + 1) / 2) # apply the cutoff function to the squared exponential partial kernels se_jm = se_jm*cut_jm k = T.sum(se_jm) # -------------------------------------------------- # FINAL FUNCTIONS # -------------------------------------------------- # energy energy kernel k_ee_fun = function([r1, r2, rho1, rho2, sig, theta, rc], k, allow_input_downcast=False, on_unused_input='warn') # energy force kernel - Used to predict energies from forces k_ef = T.grad(k, r2) k_ef_fun = function([r1, r2, rho1, rho2, sig, theta, rc], k_ef, allow_input_downcast=False, on_unused_input='warn') # force force kernel - it uses only local atom pairs to avoid useless computation k_ff = T.grad(k, r1) k_ff_der, updates = scan(lambda j, k_ff, r2: T.grad(k_ff[j], r2), sequences=T.arange(k_ff.shape[0]), non_sequences=[k_ff, r2]) k_ff_fun = function([r1, r2, rho1, rho2, sig, theta, rc], k_ff_der, allow_input_downcast=False, on_unused_input='warn') # Save the function that we want to use for multiprocessing # This is necessary because theano is a crybaby and does not want to access the # Automaticallly stored compiled object from different processes with open(Mffpath / 'k2_ee_s.pickle', 'wb') as f: pickle.dump(k_ee_fun, f) with open(Mffpath / 'k2_ef_s.pickle', 'wb') as f: pickle.dump(k_ef_fun, f) with open(Mffpath / 'k2_ff_s.pickle', 'wb') as f: pickle.dump(k_ff_fun, f) else: print("Loading Kernels") with open(Mffpath / "k2_ee_s.pickle", 'rb') as f: k_ee_fun = pickle.load(f) with open(Mffpath / "k2_ef_s.pickle", 'rb') as f: k_ef_fun = pickle.load(f) with open(Mffpath / "k2_ff_s.pickle", 'rb') as f: k_ff_fun = pickle.load(f) # -------------------------------------------------- # WRAPPERS (we don't want to plug the position of the central element every time) # -------------------------------------------------- def k2_ee(conf1, conf2, sig, theta, rc): """ Two body kernel for global energy-energy correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (float): scalar valued energy-energy 2-body kernel """ return k_ee_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def k2_ef(conf1, conf2, sig, theta, rc): """ Two body kernel for global energy-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (array): 3x1 energy-force 2-body kernel """ return -k_ef_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def k2_ff(conf1, conf2, sig, theta, rc): """ Two body kernel for force-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (matrix): 3x3 force-force 2-body kernel """ return k_ff_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) logger.info( "Ended compilation of theano two body single species kernels") return k2_ee, k2_ef, k2_ff class TwoBodyManySpeciesKernel(BaseTwoBody): """Two body many species kernel. Args: theta[0] (float): lengthscale of the kernel theta[1] (float): decay rate of the cutoff function theta[2] (float): cutoff radius """ def __init__(self, theta=(1., 1., 1.), bounds=((1e-2, 1e2), (1e-2, 1e2), (1e-2, 1e2))): super().__init__(kernel_name='TwoBodyManySpecies', theta=theta, bounds=bounds) self.type = "multi" @staticmethod def compile_theano(): """ This function generates theano compiled kernels for global energy and force learning The position of the atoms relative to the central one, and their chemical species are defined by a matrix of dimension Mx5 here called r1 and r2. Returns: k2_ee (func): energy-energy kernel k2_ef (func): energy-force kernel k2_ff (func): force-force kernel """ if not (os.path.exists(Mffpath / 'k2_ee_m.pickle') and os.path.exists(Mffpath / 'k2_ef_m.pickle') and os.path.exists(Mffpath / 'k2_ff_m.pickle')): print("Building Kernels") import theano.tensor as T from theano import function, scan logger.info("Started compilation of theano two body kernels") # -------------------------------------------------- # INITIAL DEFINITIONS # -------------------------------------------------- # positions of central atoms r1, r2 = T.dvectors('r1d', 'r2d') # positions of neighbours rho1, rho2 = T.dmatrices('rho1', 'rho2') # lengthscale hyperparameter sig = T.dscalar('sig') # cutoff hyperparameters theta = T.dscalar('theta') rc = T.dscalar('rc') # positions of neighbours without chemical species (3D space assumed) rho1s = rho1[:, 0:3] rho2s = rho2[:, 0:3] alpha_1 = rho1[:, 3] # .flatten() alpha_2 = rho2[:, 3] # .flatten() alpha_j = rho1[:, 4] # .flatten() alpha_m = rho2[:, 4] # .flatten() # numerical kronecker def delta_alpha2(a1j, a2m): d = T.exp(-(a1j - a2m) ** 2 / (2 * 1e-5 ** 2)) return d # matrices determining whether couples of atoms have the same atomic number delta_alphas12 = delta_alpha2(alpha_1[:, None], alpha_2[None, :]) delta_alphasjm = delta_alpha2(alpha_j[:, None], alpha_m[None, :]) delta_alphas1m = delta_alpha2(alpha_1[:, None], alpha_m[None, :]) delta_alphasj2 = delta_alpha2(alpha_j[:, None], alpha_2[None, :]) # distances of atoms wrt to the central one and wrt each other in 1 and 2 r1j = T.sqrt(T.sum((rho1s[:, :] - r1[None, :]) ** 2, axis=1)) r2m = T.sqrt(T.sum((rho2s[:, :] - r2[None, :]) ** 2, axis=1)) # Get the squared exponential kernels se_jm = T.exp(-(r1j[:, None] - r2m[None, :]) ** 2 / (2 * sig ** 2)) # Define cutoff function cut_jm = 0.5*(1+T.cos(np.pi*r1j[:, None]/rc))*0.5*(1+T.cos(np.pi*r2m[None, :]/rc))*( (T.sgn(rc-r1j) + 1) / 2)*((T.sgn(rc-r2m) + 1) / 2) # Apply cutoffs and chemical species masks se_jm = se_jm*cut_jm * \ (delta_alphas12 * delta_alphasjm + delta_alphas1m * delta_alphasj2) ker = T.sum(se_jm) # -------------------------------------------------- # FINAL FUNCTIONS # -------------------------------------------------- # global energy energy kernel k_ee_fun = function([r1, r2, rho1, rho2, sig, theta, rc], ker, allow_input_downcast=False, on_unused_input='warn') # energy force kernel - Used to predict energies from forces k_ef = T.grad(ker, r2) k_ef_fun = function([r1, r2, rho1, rho2, sig, theta, rc], k_ef, allow_input_downcast=False, on_unused_input='warn') # force force kernel - it uses only local atom pairs to avoid useless computation k_ff = T.grad(ker, r1) k_ff_der, updates = scan(lambda j, k_ff, r2: T.grad(k_ff[j], r2), sequences=T.arange(k_ff.shape[0]), non_sequences=[k_ff, r2]) k_ff_fun = function([r1, r2, rho1, rho2, sig, theta, rc], k_ff_der, allow_input_downcast=False, on_unused_input='warn') # Save the function that we want to use for multiprocessing # This is necessary because theano is a crybaby and does not want to access the # Automaticallly stored compiled object from different processes with open(Mffpath / 'k2_ee_m.pickle', 'wb') as f: pickle.dump(k_ee_fun, f) with open(Mffpath / 'k2_ef_m.pickle', 'wb') as f: pickle.dump(k_ef_fun, f) with open(Mffpath / 'k2_ff_m.pickle', 'wb') as f: pickle.dump(k_ff_fun, f) else: print("Loading Kernels") with open(Mffpath / "k2_ee_m.pickle", 'rb') as f: k_ee_fun = pickle.load(f) with open(Mffpath / "k2_ef_m.pickle", 'rb') as f: k_ef_fun = pickle.load(f) with open(Mffpath / "k2_ff_m.pickle", 'rb') as f: k_ff_fun = pickle.load(f) # # -------------------------------------------------- # # WRAPPERS (we don't want to plug the position of the central element every time) # # -------------------------------------------------- def k2_ee(conf1, conf2, sig, theta, rc): """ Two body kernel for global energy-energy correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (float): scalar valued energy-energy 2-body kernel """ return k_ee_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def k2_ef(conf1, conf2, sig, theta, rc): """ Two body kernel for global energy-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (array): 3x1 energy-force 2-body kernel """ return -k_ef_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def k2_ff(conf1, conf2, sig, theta, rc): """ Two body kernel for energy-energy correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (matrix): 3x3 force-force 2-body kernel """ return k_ff_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) logger.info("Ended compilation of theano two body kernels") return k2_ee, k2_ef, k2_ff
36,239
39.995475
121
py
mff
mff-master/mff/kernels/threebodykernel.py
# -*- coding: utf-8 -*- import logging import os.path import pickle from abc import ABCMeta, abstractmethod import numpy as np from mff.kernels.base import Kernel, Mffpath logger = logging.getLogger(__name__) def dummy_calc_ff(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype = data if kertype == "single": with open(Mffpath / "k3_ff_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "k3_ff_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros((len(array), 3, 3)) for i in range(len(array)): result[i] = fun(np.zeros(3), np.zeros(3), array[i][0], array[i][1], theta0, theta1, theta2) return result def dummy_calc_ee(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype, mapping = data if kertype == "single": with open(Mffpath / "k3_ee_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "k3_ee_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros(len(array)) if not mapping: for i in range(len(array)): for conf1 in array[i][0]: for conf2 in array[i][1]: result[i] += 1/9.0*fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) else: for i in range(len(array)): for conf2 in array[i][1]: result[i] += 1/3.0*fun(np.zeros(3), np.zeros(3), array[i][0], conf2, theta0, theta1, theta2) return result def dummy_calc_ef(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype, mapping = data if kertype == "single": with open(Mffpath / "k3_ef_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "k3_ef_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros((len(array), 3)) if not mapping: for i in range(len(array)): conf2 = np.array(array[i][1], dtype='float') for conf1 in array[i][0]: conf1 = np.array(conf1, dtype='float') result[i] += -1/3.0*fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) else: for i in range(len(array)): conf2 = np.array(array[i][1], dtype='float') conf1 = np.array(array[i][0], dtype='float') result[i] += -fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) return result class BaseThreeBody(Kernel, metaclass=ABCMeta): """ Three body kernel class Handles the functions common to the single-species and multi-species three-body kernels. Args: kernel_name (str): To choose between single- and two-species kernel theta[0] (float) : lengthscale of the kernel theta[1] (float) : decay rate of the cutoff function theta[2] (float) : cutoff radius bounds (list) : bounds of the kernel function. Attributes: k3_ee (object): Energy-energy kernel function k3_ef (object): Energy-force kernel function k3_ef_loc (object): Local Energy-force kernel function k3_ff (object): Force-force kernel function """ @abstractmethod def __init__(self, kernel_name, theta, bounds): super().__init__(kernel_name) self.theta = theta self.bounds = bounds self.k3_ee, self.k3_ef, self.k3_ff = self.compile_theano() def calc(self, X1, X2, ncores=1): """ Calculate the energy-force kernel between two sets of configurations. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N2*3 matrix of the vector-valued kernels """ ker = np.zeros((len(X1) * 3, len(X2) * 3)) if ncores > 1: confs = [[x1, x2] for x1 in X1 for x2 in X2] n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 3-body force-force kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in range(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in range(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores del confs import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ff, clist) del clist pool.close() pool.join() result = np.concatenate(result).reshape((n, 3, 3)) for i in range(len(X1)): for j in range(len(X2)): ker[i * 3: i * 3 + 3, 3 * j:3 * j + 3] = result[(j + i * len(X2))] del result else: for i, conf1 in enumerate(X1): for j, conf2 in enumerate(X2): ker[i * 3:i * 3 + 3, 3 * j:3 * j + 3] += self.k3_ff( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_ef(self, X_glob, X, ncores=1, mapping=False): """ Calculate the energy-force kernel between two sets of configurations. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N2*3 matrix of the vector-valued kernels """ ker = np.zeros((len(X_glob), len(X) * 3)) if ncores > 1: confs = [[x1, x2] for x1 in X_glob for x2 in X] n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 3-body energy-force kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in range(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, mapping] for i in range(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores del confs import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ef, clist) del clist pool.close() pool.join() result = np.vstack(np.asarray(result)) for i in range(len(X_glob)): for j in range(len(X)): ker[i, 3 * j:3 * j + 3] = result[(j + i * len(X))] else: if not mapping: for i, x1 in enumerate(X_glob): for j, conf2 in enumerate(X): for conf1 in x1: ker[i, 3 * j:3 * j + 3] += 1/3.0*self.k3_ef( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) else: for i, conf1 in enumerate(X_glob): for j, conf2 in enumerate(X): ker[i, 3 * j:3 * j + 3] += self.k3_ef( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_ee(self, X1, X2, ncores=1, mapping=False): """ Calculate the energy-energy kernel between two global environments. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N1 x N2 matrix of the scalar-valued kernels """ if ncores > 1: # Used for multiprocessing # Build a list of all input pairs which matrix needs to be computed confs = [[x1, x2] for x1 in X1 for x2 in X2] n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 3-body energy-energy kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in range(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, mapping] for i in range(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores del confs import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ee, clist) del clist pool.close() pool.join() result = np.concatenate(result).ravel() ker = np.zeros((len(X1), len(X2))) for i in range(len(X1)): for j in range(len(X2)): ker[i, j] = result[j + i*len(X2)] del result else: if not mapping: ker = np.zeros((len(X1), len(X2))) for i, x1 in enumerate(X1): for j, x2 in enumerate(X2): for conf1 in x1: for conf2 in x2: ker[i, j] += 1/9.0*self.k3_ee(conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) else: ker = np.zeros((len(X1), len(X2))) for i, conf1 in enumerate(X1): for j, x2 in enumerate(X2): for conf2 in x2: ker[i, j] += 1/3.0*self.k3_ee(conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_gram(self, X, ncores=1, eval_gradient=False): """ Calculate the force-force gram matrix for a set of configurations X. Args: X (list): list of N Mx5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N*3 x N*3 gram matrix of the matrix-valued kernels """ if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: confs = [[X[i], X[j]] for i in range(len(X)) for j in range(i + 1)] n = len(confs) logger.info( 'Using %i cores for the 3-body force-force gram matrix calculation' % (ncores)) import sys sys.setrecursionlimit(100000) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in range(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in range(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores del confs import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ff, clist) del clist pool.close() pool.join() result = np.concatenate(result).reshape((n, 3, 3)) off_diag = np.zeros((len(X) * 3, len(X) * 3)) diag = np.zeros((len(X) * 3, len(X) * 3)) for i in range(len(X)): diag[3 * i:3 * i + 3, 3 * i:3 * i + 3] = result[i + i * (i + 1) // 2] for j in range(i): off_diag[3 * i:3 * i + 3, 3 * j:3 * j + 3] = result[j + i * (i + 1) // 2] del result else: diag = np.zeros((X.shape[0] * 3, X.shape[0] * 3)) off_diag = np.zeros((X.shape[0] * 3, X.shape[0] * 3)) for i in range(X.shape[0]): diag[3 * i:3 * i + 3, 3 * i:3 * i + 3] = \ self.k3_ff(X[i], X[i], self.theta[0], self.theta[1], self.theta[2]) for j in range(i): off_diag[3 * i:3 * i + 3, 3 * j:3 * j + 3] = \ self.k3_ff(X[i], X[j], self.theta[0], self.theta[1], self.theta[2]) gram = diag + off_diag + off_diag.T del diag, off_diag return gram def calc_gram_e(self, X, ncores=1, eval_gradient=False): # Untested """ Calculate the energy-energy gram matrix for a set of configurations X. Args: X (list): list of N Mx5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N x N gram matrix of the scalar-valued kernels """ if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: # Build a list of all input pairs which matrix needs to be computed confs = [[X[i], X[j]] for i in range(len(X)) for j in range(i + 1)] n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 3-body energy-energy gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in range(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, False] for i in range(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores del confs import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ee, clist) del clist pool.close() pool.join() result = np.concatenate(result).ravel() off_diag = np.zeros((len(X), len(X))) diag = np.zeros((len(X), len(X))) for i in range(len(X)): diag[i, i] = result[i + i * (i + 1) // 2] for j in range(i): off_diag[i, j] = result[j + i * (i + 1) // 2] del result else: diag = np.zeros((X.shape[0], X.shape[0])) off_diag = np.zeros((X.shape[0], X.shape[0])) for i in range(X.shape[0]): for k, conf1 in enumerate(X[i]): diag[i, i] += 1/9.0*self.k3_ee(conf1, conf1, self.theta[0], self.theta[1], self.theta[2]) for conf2 in X[i][:k]: # *2 here to speed up the loop diag[i, i] += 1/9.0*2.0*self.k3_ee( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) for j in range(i): for conf1 in X[i]: for conf2 in X[j]: off_diag[i, j] += 1/9.0*self.k3_ee( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) gram = diag + off_diag + off_diag.T del diag, off_diag return gram def calc_gram_ef(self, X, X_glob, ncores=1, eval_gradient=False): """ Calculate the energy-force gram matrix for a set of configurations X. This returns a non-symmetric matrix which is equal to the transpose of the force-energy gram matrix. Args: X (list): list of N1 M1x5 arrays containing xyz coordinates and atomic species X_glob (list): list of N2 M2x5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N2 x N1*3 gram matrix of the vector-valued kernels """ gram = np.zeros((X_glob.shape[0], X.shape[0] * 3)) if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: # Multiprocessing confs = [[x1, x2] for x1 in X_glob for x2 in X] n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 3-body energy-force gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in range(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, False] for i in range(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores del confs import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ef, clist) del clist pool.close() pool.join() result = np.concatenate(result).ravel() for i in range(X_glob.shape[0]): for j in range(X.shape[0]): gram[i, 3 * j:3 * j + 3] = result[3 * (j + i * X.shape[0]):3 + 3*(j + i * X.shape[0])] del result else: for i in range(X_glob.shape[0]): for j in range(X.shape[0]): for k in X_glob[i]: gram[i, 3 * j:3 * j + 3] += 1/3.0*self.k3_ef( k, X[j], self.theta[0], self.theta[1], self.theta[2]) return gram @staticmethod @abstractmethod def compile_theano(): return None, None, None class ThreeBodySingleSpeciesKernel(BaseThreeBody): """Three body two species kernel. Args: theta[0] (float): lengthscale of the kernel theta[1] (float): decay rate of the cutoff function theta[2] (float): cutoff radius """ def __init__(self, theta=(1., 1., 1.), bounds=((1e-2, 1e2), (1e-2, 1e2), (1e-2, 1e2))): super().__init__(kernel_name='ThreeBodySingleSpecies', theta=theta, bounds=bounds) self.type = "single" @staticmethod def compile_theano(): """ This function generates theano compiled kernels for energy and force learning ker_jkmn_withcutoff = ker_jkmn #* cutoff_ikmn The position of the atoms relative to the centrla one, and their chemical species are defined by a matrix of dimension Mx5 Returns: k3_ee (func): energy-energy kernel k3_ef (func): energy-force kernel k3_ff (func): force-force kernel """ if not (os.path.exists(Mffpath / 'k3_ee_s.pickle') and os.path.exists(Mffpath / 'k3_ef_s.pickle') and os.path.exists(Mffpath / 'k3_ff_s.pickle')): print("Building Kernels") import theano.tensor as T from theano import function, scan logger.info("Started compilation of theano three body kernels") # -------------------------------------------------- # INITIAL DEFINITIONS # -------------------------------------------------- # positions of central atoms r1, r2 = T.dvectors('r1d', 'r2d') # positions of neighbours rho1, rho2 = T.dmatrices('rho1', 'rho2') # hyperparameter sig = T.dscalar('sig') # cutoff hyperparameters theta = T.dscalar('theta') rc = T.dscalar('rc') # positions of neighbours without chemical species rho1s = rho1[:, 0:3] rho2s = rho2[:, 0:3] # -------------------------------------------------- # RELATIVE DISTANCES TO CENTRAL VECTOR AND BETWEEN NEIGHBOURS # -------------------------------------------------- # first and second configuration r1j = T.sqrt(T.sum((rho1s[:, :] - r1[None, :]) ** 2, axis=1)) r2m = T.sqrt(T.sum((rho2s[:, :] - r2[None, :]) ** 2, axis=1)) rjk = T.sqrt( T.sum((rho1s[None, :, :] - rho1s[:, None, :]) ** 2, axis=2)) rmn = T.sqrt( T.sum((rho2s[None, :, :] - rho2s[:, None, :]) ** 2, axis=2)) # -------------------------------------------------- # BUILD THE KERNEL # -------------------------------------------------- # Squared exp of differences se_1j2m = T.exp(-(r1j[:, None] - r2m[None, :]) ** 2 / (2 * sig ** 2)) se_jkmn = T.exp(-(rjk[:, :, None, None] - rmn[None, None, :, :]) ** 2 / (2 * sig ** 2)) se_jk2m = T.exp(-(rjk[:, :, None] - r2m[None, None, :]) ** 2 / (2 * sig ** 2)) se_1jmn = T.exp(-(r1j[:, None, None] - rmn[None, :, :]) ** 2 / (2 * sig ** 2)) # Kernel not summed (cyclic permutations) k1n = (se_1j2m[:, None, :, None] * se_1j2m[None, :, None, :] * se_jkmn) k2n = (se_1jmn[:, None, :, :] * se_jk2m[:, :, None, :] * se_1j2m[None, :, :, None]) k3n = (se_1j2m[:, None, None, :] * se_jk2m[:, :, :, None] * se_1jmn[None, :, :, :]) # final shape is M1 M1 M2 M2 ker = k1n + k2n + k3n cut_j = 0.5*(1+T.cos(np.pi*r1j/rc))*((T.sgn(rc-r1j) + 1) / 2) cut_m = 0.5*(1+T.cos(np.pi*r2m/rc))*((T.sgn(rc-r2m) + 1) / 2) cut_jk = cut_j[:,None]*cut_j[None,:]*0.5*(1+T.cos(np.pi*rjk/rc))*((T.sgn(rc-rjk) + 1) / 2) cut_mn = cut_m[:,None]*cut_m[None,:]*0.5*(1+T.cos(np.pi*rmn/rc))*((T.sgn(rc-rmn) + 1) / 2) # -------------------------------------------------- # REMOVE DIAGONAL ELEMENTS AND ADD CUTOFF # -------------------------------------------------- # remove diagonal elements AND lower triangular ones from first configuration mask_jk = T.triu(T.ones_like(rjk)) - T.identity_like(rjk) # remove diagonal elements from second configuration mask_mn = T.ones_like(rmn) - T.identity_like(rmn) # Combine masks mask_jkmn = mask_jk[:, :, None, None] * mask_mn[None, None, :, :] # Apply mask and then apply cutoff functions ker = ker * mask_jkmn ker = T.sum(ker * cut_jk[:, :, None, None] * cut_mn[None, None, :, :]) # -------------------------------------------------- # FINAL FUNCTIONS # -------------------------------------------------- # global energy energy kernel k_ee_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], ker, on_unused_input='ignore') # global energy force kernel k_ef = T.grad(ker, r2) k_ef_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], k_ef, on_unused_input='ignore') # local force force kernel k_ff = T.grad(ker, r1) k_ff_der, updates = scan(lambda j, k_ff, r2: T.grad(k_ff[j], r2), sequences=T.arange(k_ff.shape[0]), non_sequences=[k_ff, r2]) k_ff_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], k_ff_der, on_unused_input='ignore') # Save the function that we want to use for multiprocessing # This is necessary because theano is a crybaby and does not want to access the # Automaticallly stored compiled object from different processes with open(Mffpath / 'k3_ee_s.pickle', 'wb') as f: pickle.dump(k_ee_fun, f) with open(Mffpath / 'k3_ef_s.pickle', 'wb') as f: pickle.dump(k_ef_fun, f) with open(Mffpath / 'k3_ff_s.pickle', 'wb') as f: pickle.dump(k_ff_fun, f) else: print("Loading Kernels") with open(Mffpath / "k3_ee_s.pickle", 'rb') as f: k_ee_fun = pickle.load(f) with open(Mffpath / "k3_ef_s.pickle", 'rb') as f: k_ef_fun = pickle.load(f) with open(Mffpath / "k3_ff_s.pickle", 'rb') as f: k_ff_fun = pickle.load(f) # WRAPPERS (we don't want to plug the position of the central element every time) def k3_ee(conf1, conf2, sig, theta, rc): """ Three body kernel for global energy-energy correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (float): scalar valued energy-energy 3-body kernel """ return k_ee_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def k3_ef(conf1, conf2, sig, theta, rc): """ Three body kernel for global energy-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (array): 3x1 energy-force 3-body kernel """ return -k_ef_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def k3_ff(conf1, conf2, sig, theta, rc): """ Three body kernel for local force-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (matrix): 3x3 force-force 3-body kernel """ return k_ff_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) logger.info("Ended compilation of theano three body kernels") return k3_ee, k3_ef, k3_ff class ThreeBodyManySpeciesKernel(BaseThreeBody): """Three body many species kernel. Args: theta[0] (float): lengthscale of the kernel theta[1] (float): decay rate of the cutoff function theta[2] (float): cutoff radius """ def __init__(self, theta=(1., 1., 1.), bounds=((1e-2, 1e2), (1e-2, 1e2), (1e-2, 1e2))): super().__init__(kernel_name='ThreeBodyManySpecies', theta=theta, bounds=bounds) self.type = "multi" @staticmethod def compile_theano(): """ This function generates theano compiled kernels for energy and force learning ker_jkmn_withcutoff = ker_jkmn #* cutoff_ikmn The position of the atoms relative to the centrla one, and their chemical species are defined by a matrix of dimension Mx5 Returns: k3_ee (func): energy-energy kernel k3_ef (func): energy-force kernel k3_ff (func): force-force kernel """ logger.info("Started compilation of theano three body kernels") if not (os.path.exists(Mffpath / 'k3_ee_m.pickle') and os.path.exists(Mffpath / 'k3_ef_m.pickle') and os.path.exists(Mffpath / 'k3_ff_m.pickle')): print("Building Kernels") import theano.tensor as T from theano import function, scan # -------------------------------------------------- # INITIAL DEFINITIONS # -------------------------------------------------- # positions of central atoms r1, r2 = T.dvectors('r1d', 'r2d') # positions of neighbours rho1, rho2 = T.dmatrices('rho1', 'rho2') # hyperparameter sig = T.dscalar('sig') # cutoff hyperparameters theta = T.dscalar('theta') rc = T.dscalar('rc') # positions of neighbours without chemical species rho1s = rho1[:, 0:3] rho2s = rho2[:, 0:3] alpha_1 = rho1[:, 3].flatten() alpha_2 = rho2[:, 3].flatten() alpha_j = rho1[:, 4].flatten() alpha_m = rho2[:, 4].flatten() alpha_k = rho1[:, 4].flatten() alpha_n = rho2[:, 4].flatten() # -------------------------------------------------- # RELATIVE DISTANCES TO CENTRAL VECTOR AND BETWEEN NEIGHBOURS # -------------------------------------------------- # first and second configuration r1j = T.sqrt(T.sum((rho1s[:, :] - r1[None, :]) ** 2, axis=1)) r2m = T.sqrt(T.sum((rho2s[:, :] - r2[None, :]) ** 2, axis=1)) rjk = T.sqrt( T.sum((rho1s[None, :, :] - rho1s[:, None, :]) ** 2, axis=2)) rmn = T.sqrt( T.sum((rho2s[None, :, :] - rho2s[:, None, :]) ** 2, axis=2)) # -------------------------------------------------- # CHEMICAL SPECIES MASK # -------------------------------------------------- # numerical kronecker def delta_alpha2(a1j, a2m): d = np.exp(-(a1j - a2m) ** 2 / (2 * 0.00001 ** 2)) return d # permutation 1 delta_alphas12 = delta_alpha2(alpha_1[0], alpha_2[0]) delta_alphasjm = delta_alpha2(alpha_j[:, None], alpha_m[None, :]) delta_alphas_jmkn = delta_alphasjm[:, None, :, None] * delta_alphasjm[None, :, None, :] delta_perm1 = delta_alphas12 * delta_alphas_jmkn # permutation 3 delta_alphas1m = delta_alpha2( alpha_1[0, None], alpha_m[None, :]).flatten() delta_alphasjn = delta_alpha2(alpha_j[:, None], alpha_n[None, :]) delta_alphask2 = delta_alpha2( alpha_k[:, None], alpha_2[None, 0]).flatten() delta_perm3 = delta_alphas1m[None, None, :, None] * delta_alphasjn[:, None, None, :] * \ delta_alphask2[None, :, None, None] # permutation 5 delta_alphas1n = delta_alpha2( alpha_1[0, None], alpha_n[None, :]).flatten() delta_alphasj2 = delta_alpha2( alpha_j[:, None], alpha_2[None, 0]).flatten() delta_alphaskm = delta_alpha2(alpha_k[:, None], alpha_m[None, :]) delta_perm5 = delta_alphas1n[None, None, None, :] * delta_alphaskm[None, :, :, None] * \ delta_alphasj2[:, None, None, None] # -------------------------------------------------- # BUILD THE KERNEL # -------------------------------------------------- # Squared exp of differences se_1j2m = T.exp(-(r1j[:, None] - r2m[None, :]) ** 2 / (2 * sig ** 2)) se_jkmn = T.exp(-(rjk[:, :, None, None] - rmn[None, None, :, :]) ** 2 / (2 * sig ** 2)) se_jk2m = T.exp(-(rjk[:, :, None] - r2m[None, None, :]) ** 2 / (2 * sig ** 2)) se_1jmn = T.exp(-(r1j[:, None, None] - rmn[None, :, :]) ** 2 / (2 * sig ** 2)) # Kernel not summed (cyclic permutations) k1n = (se_1j2m[:, None, :, None] * se_1j2m[None, :, None, :] * se_jkmn) k2n = (se_1jmn[:, None, :, :] * se_jk2m[:, :, None, :] * se_1j2m[None, :, :, None]) k3n = (se_1j2m[:, None, None, :] * se_jk2m[:, :, :, None] * se_1jmn[None, :, :, :]) # final shape is M1 M1 M2 M2 ker_loc = k1n * delta_perm1 + k2n * delta_perm3 + k3n * delta_perm5 # Faster version of cutoff (less calculations) cut_j = 0.5*(1+T.cos(np.pi*r1j/rc))#*((T.sgn(rc-r1j) + 1) / 2) cut_m = 0.5*(1+T.cos(np.pi*r2m/rc))#*((T.sgn(rc-r2m) + 1) / 2) cut_jk = cut_j[:,None]*cut_j[None,:]*0.5*(1+T.cos(np.pi*rjk/rc))*((T.sgn(rc-rjk) + 1) / 2) cut_mn = cut_m[:,None]*cut_m[None,:]*0.5*(1+T.cos(np.pi*rmn/rc))*((T.sgn(rc-rmn) + 1) / 2) # -------------------------------------------------- # REMOVE DIAGONAL ELEMENTS # -------------------------------------------------- # remove diagonal elements AND lower triangular ones from first configuration mask_jk = T.triu(T.ones_like(rjk)) - T.identity_like(rjk) # remove diagonal elements from second configuration mask_mn = T.ones_like(rmn) - T.identity_like(rmn) # Combine masks mask_jkmn = mask_jk[:, :, None, None] * mask_mn[None, None, :, :] # Apply mask and then apply cutoff functions ker_loc = ker_loc * mask_jkmn ker_loc = T.sum( ker_loc * cut_jk[:, :, None, None] * cut_mn[None, None, :, :]) # -------------------------------------------------- # FINAL FUNCTIONS # -------------------------------------------------- # energy energy kernel k_ee_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], ker_loc, on_unused_input='ignore') # energy force kernel k_ef_cut = T.grad(ker_loc, r2) k_ef_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], k_ef_cut, on_unused_input='ignore') # force force kernel k_ff_cut = T.grad(ker_loc, r1) k_ff_cut_der, updates = scan(lambda j, k_ff_cut, r2: T.grad(k_ff_cut[j], r2), sequences=T.arange(k_ff_cut.shape[0]), non_sequences=[k_ff_cut, r2]) k_ff_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], k_ff_cut_der, on_unused_input='ignore') # Save the function that we want to use for multiprocessing # This is necessary because theano is a crybaby and does not want to access the # Automaticallly stored compiled object from different processes with open(Mffpath / 'k3_ee_m.pickle', 'wb') as f: pickle.dump(k_ee_fun, f) with open(Mffpath / 'k3_ef_m.pickle', 'wb') as f: pickle.dump(k_ef_fun, f) with open(Mffpath / 'k3_ff_m.pickle', 'wb') as f: pickle.dump(k_ff_fun, f) else: print("Loading Kernels") with open(Mffpath / "k3_ee_m.pickle", 'rb') as f: k_ee_fun = pickle.load(f) with open(Mffpath / "k3_ef_m.pickle", 'rb') as f: k_ef_fun = pickle.load(f) with open(Mffpath / "k3_ff_m.pickle", 'rb') as f: k_ff_fun = pickle.load(f) # WRAPPERS (we don't want to plug the position of the central element every time) def k3_ee(conf1, conf2, sig, theta, rc): """ Three body kernel for energy-energy correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (float): scalar valued energy-energy 3-body kernel """ return k_ee_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def k3_ef(conf1, conf2, sig, theta, rc): """ Three body kernel for energy-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (array): 3x1 energy-force 3-body kernel """ return -k_ef_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def k3_ff(conf1, conf2, sig, theta, rc): """ Three body kernel for force-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (matrix): 3x3 force-force 3-body kernel """ return k_ff_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) logger.info("Ended compilation of theano three body kernels") return k3_ee, k3_ef, k3_ff
40,850
39.728814
117
py
mff
mff-master/mff/kernels/eamkernel.py
# -*- coding: utf-8 -*- import logging import os.path import pickle from abc import ABCMeta, abstractmethod import numpy as np from mff.kernels.base import Kernel, Mffpath logger = logging.getLogger(__name__) def dummy_calc_ff(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype = data if kertype == "single": with open(Mffpath / "keam_ff_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "keam_ff_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros((len(array), 3, 3)) for i in np.arange(len(array)): result[i] = fun(np.zeros(3), np.zeros(3), array[i][0], array[i][1], theta0, theta1, theta2) return result def dummy_calc_ee(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype, mapping, alpha_1_descr = data if mapping: if kertype == "single": with open(Mffpath / "keam_eed_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "keam_eed_m.pickle", 'rb') as f: fun = pickle.load(f) else: if kertype == "single": with open(Mffpath / "keam_ee_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "keam_ee_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros(len(array)) if not mapping: for i in np.arange(len(array)): for conf1 in array[i][0]: for conf2 in array[i][1]: result[i] += 0.25*fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) else: if kertype == "multi": for i in np.arange(len(array)): for conf2 in array[i][1]: result[i] += 0.5*fun(np.zeros(3), array[i] [0], conf2, theta0, theta1, theta2, alpha_1_descr) else: for i in np.arange(len(array)): for conf2 in array[i][1]: result[i] += fun(np.zeros(3), array[i] [0], conf2, theta0, theta1, theta2) return result def dummy_calc_ef(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype, mapping, alpha_1_descr = data if mapping: if kertype == "single": with open(Mffpath / "keam_efd_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "keam_efd_m.pickle", 'rb') as f: fun = pickle.load(f) else: if kertype == "single": with open(Mffpath / "keam_ef_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "keam_ef_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros((len(array), 3)) if not mapping: for i in np.arange(len(array)): conf2 = np.array(array[i][1], dtype='float') for conf1 in array[i][0]: conf1 = np.array(conf1, dtype='float') result[i] += -0.5*fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) else: if kertype == "multi": for i in np.arange(len(array)): conf2 = np.array(array[i][1], dtype='float') conf1 = np.array(array[i][0], dtype='float') result[i] += -fun(np.zeros(3), conf1, conf2, theta0, theta1, theta2, alpha_1_descr) else: for i in np.arange(len(array)): conf2 = np.array(array[i][1], dtype='float') conf1 = np.array(array[i][0], dtype='float') result[i] += -fun(np.zeros(3), conf1, conf2, theta0, theta1, theta2) return result class BaseEam(Kernel, metaclass=ABCMeta): """ Eam kernel class Handles the functions common to the single-species and multi-species two-body kernels. Args: kernel_name (str): To choose between single- and two-species kernel theta[0] (float) : lengthscale of the kernel theta[1] (float) : decay rate of the cutoff function theta[2] (float) : cutoff radius bounds (list) : bounds of the kernel function. Attributes: k2_ee (object): Energy-energy kernel function k2_ef (object): Energy-force kernel function k2_ff (object): Force-force kernel function """ @abstractmethod def __init__(self, kernel_name, theta, bounds): super().__init__(kernel_name) self.theta = theta self.bounds = bounds self.k2_ee, self.k2_ef, self.k2_ff, self.k2_ee_d, self.k2_ef_d = self.compile_theano() def calc(self, X1, X2, ncores=1): """ Calculate the energy-force kernel between two sets of configurations. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N2*3 matrix of the vector-valued kernels """ ker = np.zeros((len(X1) * 3, len(X2) * 3)) if ncores > 1: confs = [] for x1 in X1: for x2 in X2: confs.append(np.asarray([x1, x2])) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the eam force-force kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ff, clist) pool.close() pool.join() result = np.concatenate(result).reshape((n, 3, 3)) for i in range(len(X1)): for j in range(len(X2)): ker[i * 3: i * 3 + 3, 3 * j:3 * j + 3] = result[(j + i * len(X2))] else: for i, conf1 in enumerate(X1): for j, conf2 in enumerate(X2): ker[i * 3:i * 3 + 3, 3 * j:3 * j + 3] += self.k2_ff( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_ef(self, X_glob, X, ncores=1, mapping=False, alpha_1_descr=0): """ Calculate the energy-force kernel between two sets of configurations. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N2*3 matrix of the vector-valued kernels """ ker = np.zeros((len(X_glob), len(X) * 3)) if ncores > 1: confs = [] for x1 in X_glob: for x2 in X: confs.append(np.asarray([x1, x2])) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the eam energy-force kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, mapping, alpha_1_descr] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ef, clist) pool.close() pool.join() result = np.vstack(np.asarray(result)) for i in range(len(X_glob)): for j in range(len(X)): ker[i, 3 * j:3 * j + 3] = result[(j + i * len(X))] else: if not mapping: for i, x1 in enumerate(X_glob): for j, conf2 in enumerate(X): for conf1 in x1: ker[i, 3 * j:3 * j + 3] += 0.5*self.k2_ef( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) else: if self.type == 'multi': for i, conf1 in enumerate(X_glob): for j, conf2 in enumerate(X): ker[i, 3 * j:3 * j + 3] += self.k2_ef_d( conf1, conf2, self.theta[0], self.theta[1], self.theta[2], alpha_1_descr) else: for i, conf1 in enumerate(X_glob): for j, conf2 in enumerate(X): ker[i, 3 * j:3 * j + 3] += self.k2_ef_d( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_ee(self, X1, X2, ncores=1, mapping=False, alpha_1_descr=0): """ Calculate the energy-energy kernel between two global environments. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N1 x N2 matrix of the scalar-valued kernels """ if ncores > 1: # Used for multiprocessing confs = [] # Build a list of all input pairs which matrix needs to be computed for x1 in X1: for x2 in X2: confs.append(np.asarray([x1, x2])) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the eam energy-energy kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, mapping, alpha_1_descr] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ee, clist) pool.close() pool.join() result = np.concatenate(result).ravel() ker = np.zeros((len(X1), len(X2))) for i in range(len(X1)): for j in range(len(X2)): ker[i, j] = result[j + i*len(X2)] else: if not mapping: ker = np.zeros((len(X1), len(X2))) for i, x1 in enumerate(X1): for j, x2 in enumerate(X2): for conf1 in x1: for conf2 in x2: ker[i, j] += 0.25*self.k2_ee(conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) else: if self.type == 'multi': ker = np.zeros((len(X1), len(X2))) for i, conf1 in enumerate(X1): for j, x2 in enumerate(X2): for conf2 in x2: ker[i, j] += 0.5*self.k2_ee_d( conf1, conf2, self.theta[0], self.theta[1], self.theta[2], alpha_1_descr) else: ker = np.zeros((len(X1), len(X2))) for i, conf1 in enumerate(X1): for j, x2 in enumerate(X2): for conf2 in x2: ker[i, j] += 0.5*self.k2_ee_d( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_gram(self, X, ncores=1, eval_gradient=False): """ Calculate the force-force gram matrix for a set of configurations X. Args: X (list): list of N Mx5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N*3 x N*3 gram matrix of the matrix-valued kernels """ if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: # Used for multiprocessing confs = [] # Build a list of all input pairs which matrix needs to be computed for i in np.arange(len(X)): for j in np.arange(i + 1): thislist = np.asarray([X[i], X[j]]) confs.append(thislist) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the eam force-force gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ff, clist) pool.close() pool.join() result = np.concatenate(result).reshape((n, 3, 3)) off_diag = np.zeros((len(X) * 3, len(X) * 3)) diag = np.zeros((len(X) * 3, len(X) * 3)) for i in np.arange(len(X)): diag[3 * i:3 * i + 3, 3 * i:3 * i + 3] = result[i + i * (i + 1) // 2] for j in np.arange(i): off_diag[3 * i:3 * i + 3, 3 * j:3 * j + 3] = result[j + i * (i + 1) // 2] else: diag = np.zeros((X.shape[0] * 3, X.shape[0] * 3)) off_diag = np.zeros((X.shape[0] * 3, X.shape[0] * 3)) for i in np.arange(X.shape[0]): diag[3 * i:3 * i + 3, 3 * i:3 * i + 3] = \ self.k2_ff(X[i], X[i], self.theta[0], self.theta[1], self.theta[2]) for j in np.arange(i): off_diag[3 * i:3 * i + 3, 3 * j:3 * j + 3] = \ self.k2_ff(X[i], X[j], self.theta[0], self.theta[1], self.theta[2]) gram = diag + off_diag + off_diag.T # The gram matrix is symmetric return gram def calc_gram_e(self, X, ncores=1, eval_gradient=False): """ Calculate the energy-energy gram matrix for a set of configurations X. Args: X (list): list of N Mx5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N x N gram matrix of the scalar-valued kernels """ if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: # Used for multiprocessing confs = [] # Build a list of all input pairs which matrix needs to be computed for i in np.arange(len(X)): for j in np.arange(i + 1): thislist = np.array([list(X[i]), list(X[j])]) confs.append(thislist) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the eam energy-energy gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, False, False] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ee, clist) pool.close() pool.join() result = np.concatenate(result).ravel() off_diag = np.zeros((len(X), len(X))) diag = np.zeros((len(X), len(X))) for i in np.arange(len(X)): diag[i, i] = result[i + i * (i + 1) // 2] for j in np.arange(i): off_diag[i, j] = result[j + i * (i + 1) // 2] else: diag = np.zeros((X.shape[0], X.shape[0])) off_diag = np.zeros((X.shape[0], X.shape[0])) for i in np.arange(X.shape[0]): for k, conf1 in enumerate(X[i]): diag[i, i] += 0.25*self.k2_ee(conf1, conf1, self.theta[0], self.theta[1], self.theta[2]) for conf2 in X[i][:k]: # *2 here to speed up the loop diag[i, i] += 0.25*2.0*self.k2_ee( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) for j in np.arange(i): for conf1 in X[i]: for conf2 in X[j]: off_diag[i, j] += 0.25*self.k2_ee( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) gram = diag + off_diag + off_diag.T # Gram matrix is symmetric return gram def calc_gram_ef(self, X, X_glob, ncores=1, eval_gradient=False): """ Calculate the energy-force gram matrix for a set of configurations X. This returns a non-symmetric matrix which is equal to the transpose of the force-energy gram matrix. Args: X (list): list of N1 M1x5 arrays containing xyz coordinates and atomic species X_glob (list): list of N2 M2x5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N2 x N1*3 gram matrix of the vector-valued kernels """ gram = np.zeros((X_glob.shape[0], X.shape[0] * 3)) if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: # Multiprocessing confs = [] for i in np.arange(len(X_glob)): for j in np.arange(len(X)): thislist = np.asarray([X_glob[i], X[j]]) confs.append(thislist) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the eam energy-force gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type, False, False] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ef, clist) pool.close() pool.join() result = np.vstack(np.asarray(result)) for i in np.arange(X_glob.shape[0]): for j in np.arange(X.shape[0]): gram[i, 3 * j:3 * j + 3] = result[(j + i * X.shape[0])] else: for i in np.arange(X_glob.shape[0]): for j in np.arange(X.shape[0]): for k in X_glob[i]: gram[i, 3 * j:3 * j + 3] += 0.5*self.k2_ef( k, X[j], self.theta[0], self.theta[1], self.theta[2]) self.gram_ef = gram return gram @staticmethod @abstractmethod def compile_theano(): return None, None, None, None, None class EamSingleSpeciesKernel(BaseEam): """Eam single species kernel. Args: theta[0] (float): lengthscale of the kernel theta[1] (float): cutoff radius theta[2] (float): radius in the descriptor's exponent """ def __init__(self, theta=(1., 1., 1.), bounds=((1e-2, 1e2), (1e-1, 1e2), (1e-1, 1e2))): super().__init__(kernel_name='EamSingleSpecies', theta=theta, bounds=bounds) self.type = "single" @staticmethod def compile_theano(): """ This function generates theano compiled kernels for global energy and force learning The position of the atoms relative to the central one, and their chemical species are defined by a matrix of dimension Mx5 here called r1 and r2. Returns: k2_ee (func): energy-energy kernel k2_ef (func): energy-force kernel k2_ff (func): force-force kernel k2_ee_map (func): energy-energy kernel that takes descriptor as one argument k2_ef_map (func): energy-force kernel that takes descriptor as one argument """ if not (os.path.exists(Mffpath / 'keam_ee_s.pickle') and os.path.exists(Mffpath / 'keam_ef_s.pickle') and os.path.exists( Mffpath / 'keam_ff_s.pickle') and os.path.exists(Mffpath / 'keam_eed_s.pickle') and os.path.exists(Mffpath / 'keam_efd_s.pickle')): print("Building Kernels") import theano.tensor as T from theano import function, scan logger.info( "Started compilation of theano eam single species kernels") # -------------------------------------------------- # INITIAL DEFINITIONS # -------------------------------------------------- # positions of central atoms r1, r2 = T.dvectors('r1d', 'r2d') # positions of neighbours rho1, rho2 = T.dmatrices('rho1', 'rho2') # lengthscale hyperparameter sig = T.dscalar('sig') # cutoff hyperparameters rc = T.dscalar('rc') # Descriptor as a given input, used to map q1_descr = T.dscalar('q1_descr') # Radius to use at denominator in the descriptor r0 = T.dscalar('r0') # positions of neighbours without chemical species (3D space assumed) rho1s = rho1[:, 0:3] rho2s = rho2[:, 0:3] # distances of atoms wrt to the central one and wrt each other in 1 and 2 r1j = T.sqrt(T.sum((rho1s[:, :] - r1[None, :]) ** 2, axis=1)) r2m = T.sqrt(T.sum((rho2s[:, :] - r2[None, :]) ** 2, axis=1)) esp_term_1 = (r1j/r0 - 1) esp_term_2 = (r2m/r0 - 1) cut_1 = 0.5*(1 + T.cos(np.pi*r1j/rc))*((T.sgn(rc-r1j) + 1) / 2) cut_2 = 0.5*(1 + T.cos(np.pi*r2m/rc))*((T.sgn(rc-r2m) + 1) / 2) q1 = T.sum(T.exp(-esp_term_1)*cut_1) q2 = T.sum(T.exp(-esp_term_2)*cut_2) k = T.exp(-(q1-q2)**2/(2*sig**2)) k_descr = T.exp(-(q1_descr-q2)**2/(2*sig**2)) # energy energy kernel k_ee_fun = function([r1, r2, rho1, rho2, sig, rc, r0], k, allow_input_downcast=False, on_unused_input='warn') # energy force kernel - Used to predict energies from forces k_ef = T.grad(k, r2) k_ef_fun = function([r1, r2, rho1, rho2, sig, rc, r0], k_ef, allow_input_downcast=False, on_unused_input='warn') # force force kernel - it uses only local atom pairs to avoid useless computation k_ff = T.grad(k, r1) k_ff_der, updates = scan(lambda j, k_ff, r2: T.grad(k_ff[j], r2), sequences=T.arange(k_ff.shape[0]), non_sequences=[k_ff, r2]) k_ff_fun = function([r1, r2, rho1, rho2, sig, rc, r0], k_ff_der, allow_input_downcast=False, on_unused_input='warn') # energy energy descriptor kernel k_ee_fun_d = function([r2, q1_descr, rho2, sig, rc, r0], k_descr, allow_input_downcast=False, on_unused_input='warn') # energy force descriptor kernel k_ef_descr = T.grad(k_descr, r2) k_ef_fun_d = function([r2, q1_descr, rho2, sig, rc, r0], k_ef_descr, allow_input_downcast=False, on_unused_input='warn') # Save the function that we want to use for multiprocessing # This is necessary because theano is a crybaby and does not want to access the # Automaticallly stored compiled object from different processes with open(Mffpath / 'keam_ee_s.pickle', 'wb') as f: pickle.dump(k_ee_fun, f) with open(Mffpath / 'keam_ef_s.pickle', 'wb') as f: pickle.dump(k_ef_fun, f) with open(Mffpath / 'keam_ff_s.pickle', 'wb') as f: pickle.dump(k_ff_fun, f) with open(Mffpath / 'keam_eed_s.pickle', 'wb') as f: pickle.dump(k_ee_fun_d, f) with open(Mffpath / 'keam_efd_s.pickle', 'wb') as f: pickle.dump(k_ef_fun_d, f) else: print("Loading Kernels") with open(Mffpath / "keam_ee_s.pickle", 'rb') as f: k_ee_fun = pickle.load(f) with open(Mffpath / "keam_ef_s.pickle", 'rb') as f: k_ef_fun = pickle.load(f) with open(Mffpath / "keam_ff_s.pickle", 'rb') as f: k_ff_fun = pickle.load(f) with open(Mffpath / 'keam_eed_s.pickle', 'rb') as f: k_ee_fun_d = pickle.load(f) with open(Mffpath / 'keam_efd_s.pickle', 'rb') as f: k_ef_fun_d = pickle.load(f) # -------------------------------------------------- # WRAPPERS (we don't want to plug the position of the central element every time) # -------------------------------------------------- def k2_ee(conf1, conf2, sig, rc, r0): """ Eam kernel for global energy-energy correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[1] Returns: kernel (float): scalar valued energy-energy Eam kernel """ return k_ee_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, rc, r0) def k2_ef(conf1, conf2, sig, rc, r0): """ Eam kernel for global energy-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[1] Returns: kernel (array): 3x1 energy-force Eam kernel """ return -k_ef_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, rc, r0) def k2_ff(conf1, conf2, sig, rc, r0): """ Eam kernel for force-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[1] Returns: kernel (matrix): 3x3 force-force Eam kernel """ return k_ff_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, rc, r0) def k2_ee_d(descr1, conf2, sig, rc, r0): """ Eam kernel for global energy-force correlation Args: descr1 (float): descriptor calculated for the first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[1] Returns: kernel (array): 3x1 energy-force Eam kernel """ return k_ee_fun_d(np.zeros(3), descr1, conf2, sig, rc, r0) def k2_ef_d(descr1, conf2, sig, rc, r0): """ Eam kernel for force-force correlation Args: descr1 (float): descriptor calculated for the first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[`] Returns: kernel (matrix): 3x3 force-force Eam kernel """ return -k_ef_fun_d(np.zeros(3), descr1, conf2, sig, rc, r0) logger.info("Ended compilation of theano eam single species kernels") return k2_ee, k2_ef, k2_ff, k2_ee_d, k2_ef_d class EamManySpeciesKernel(BaseEam): """Eam multi species kernel. Args: theta[0] (float): lengthscale of the kernel theta[1] (float): cutoff radius theta[2] (float): radius in the descriptor's exponent """ def __init__(self, theta=(1., 1., 1.), bounds=((1e-2, 1e2), (1e-2, 1e2), (1e-2, 1e2))): super().__init__(kernel_name='EamMultiSpecies', theta=theta, bounds=bounds) self.type = "multi" @staticmethod def compile_theano(): """ This function generates theano compiled kernels for global energy and force learning The position of the atoms relative to the central one, and their chemical species are defined by a matrix of dimension Mx5 here called r1 and r2. Returns: k2_ee (func): energy-energy kernel k2_ef (func): energy-force kernel k2_ff (func): force-force kernel k2_ee_map (func): energy-energy kernel that takes descriptor as one argument k2_ef_map (func): energy-force kernel that takes descriptor as one argument """ if not (os.path.exists(Mffpath / 'keam_ee_m.pickle') and os.path.exists(Mffpath / 'keam_ef_m.pickle') and os.path.exists( Mffpath / 'keam_ff_m.pickle') and os.path.exists(Mffpath / 'keam_eed_m.pickle') and os.path.exists(Mffpath / 'keam_efd_m.pickle')): print("Building Kernels") import theano.tensor as T from theano import function, scan logger.info( "Started compilation of theano eam multi species kernels") # -------------------------------------------------- # INITIAL DEFINITIONS # -------------------------------------------------- # positions of central atoms r1, r2 = T.dvectors('r1d', 'r2d') # positions of neighbours rho1, rho2 = T.dmatrices('rho1', 'rho2') # lengthscale hyperparameter sig = T.dscalar('sig') # cutoff hyperparameters rc = T.dscalar('rc') # Descriptor as a given input, used to map q1_descr = T.dscalar('q1_descr') # Element of the central atom if descriptor is Given alpha_1_descr = T.dscalar('alpha_1_descr') # Radius to use at denominator in the descriptor r0 = T.dscalar('r0') # positions of neighbours without chemical species (3D space assumed) rho1s = rho1[:, 0:3] rho2s = rho2[:, 0:3] alpha_1 = rho1[0, 3] # .flatten() alpha_2 = rho2[0, 3] # .flatten() # numerical kronecker def delta_alpha(a1j, a2m): d = T.exp(-(a1j - a2m) ** 2 / (2 * 1e-5 ** 2)) return d # matrices determining whether couples of atoms have the same atomic number delta_alpha_12 = delta_alpha(alpha_1, alpha_2) delta_alpha_12_descr = delta_alpha(alpha_1_descr, alpha_2) # distances of atoms wrt to the central one and wrt each other in 1 and 2 r1j = T.sqrt(T.sum((rho1s[:, :] - r1[None, :]) ** 2, axis=1)) r2m = T.sqrt(T.sum((rho2s[:, :] - r2[None, :]) ** 2, axis=1)) esp_term_1 = (r1j/r0 - 1) esp_term_2 = (r2m/r0 - 1) cut_1 = 0.5*(1 + T.cos(np.pi*r1j/rc))*((T.sgn(rc-r1j) + 1) / 2) cut_2 = 0.5*(1 + T.cos(np.pi*r2m/rc))*((T.sgn(rc-r2m) + 1) / 2) q1 = T.sum(T.exp(-esp_term_1)*cut_1) q2 = T.sum(T.exp(-esp_term_2)*cut_2) k = T.exp(-(q1-q2)**2/(2*sig**2))*delta_alpha_12 k_descr = T.exp(-(q1_descr-q2)**2/(2*sig**2))*delta_alpha_12_descr # energy energy kernel k_ee_fun = function([r1, r2, rho1, rho2, sig, rc, r0], k, allow_input_downcast=False, on_unused_input='warn') # energy force kernel - Used to predict energies from forces k_ef = T.grad(k, r2) k_ef_fun = function([r1, r2, rho1, rho2, sig, rc, r0], k_ef, allow_input_downcast=False, on_unused_input='warn') # force force kernel - it uses only local atom pairs to avoid useless computation k_ff = T.grad(k, r1) k_ff_der, updates = scan(lambda j, k_ff, r2: T.grad(k_ff[j], r2), sequences=T.arange(k_ff.shape[0]), non_sequences=[k_ff, r2]) k_ff_fun = function([r1, r2, rho1, rho2, sig, rc, r0], k_ff_der, allow_input_downcast=False, on_unused_input='warn') # energy energy descriptor kernel k_ee_fun_d = function([r2, q1_descr, rho2, sig, rc, r0, alpha_1_descr], k_descr, allow_input_downcast=False, on_unused_input='warn') # energy force descriptor kernel k_ef_descr = T.grad(k_descr, r2) k_ef_fun_d = function([r2, q1_descr, rho2, sig, rc, r0, alpha_1_descr], k_ef_descr, allow_input_downcast=False, on_unused_input='warn') # Save the function that we want to use for multiprocessing # This is necessary because theano is a crybaby and does not want to access the # Automaticallly stored compiled object from different processes with open(Mffpath / 'keam_ee_m.pickle', 'wb') as f: pickle.dump(k_ee_fun, f) with open(Mffpath / 'keam_ef_m.pickle', 'wb') as f: pickle.dump(k_ef_fun, f) with open(Mffpath / 'keam_ff_m.pickle', 'wb') as f: pickle.dump(k_ff_fun, f) with open(Mffpath / 'keam_eed_m.pickle', 'wb') as f: pickle.dump(k_ee_fun_d, f) with open(Mffpath / 'keam_efd_m.pickle', 'wb') as f: pickle.dump(k_ef_fun_d, f) else: print("Loading Kernels") with open(Mffpath / "keam_ee_m.pickle", 'rb') as f: k_ee_fun = pickle.load(f) with open(Mffpath / "keam_ef_m.pickle", 'rb') as f: k_ef_fun = pickle.load(f) with open(Mffpath / "keam_ff_m.pickle", 'rb') as f: k_ff_fun = pickle.load(f) with open(Mffpath / 'keam_eed_m.pickle', 'rb') as f: k_ee_fun_d = pickle.load(f) with open(Mffpath / 'keam_efd_m.pickle', 'rb') as f: k_ef_fun_d = pickle.load(f) # -------------------------------------------------- # WRAPPERS (we don't want to plug the position of the central element every time) # -------------------------------------------------- def k2_ee(conf1, conf2, sig, rc, r0): """ Eam kernel for global energy-energy correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[1] Returns: kernel (float): scalar valued energy-energy Eam kernel """ return k_ee_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, rc, r0) def k2_ef(conf1, conf2, sig, rc, r0): """ Eam kernel for global energy-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[1] Returns: kernel (array): 3x1 energy-force Eam kernel """ return -k_ef_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, rc, r0) def k2_ff(conf1, conf2, sig, rc, r0): """ Eam kernel for force-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[1] Returns: kernel (matrix): 3x3 force-force Eam kernel """ return k_ff_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, rc, r0) def k2_ee_d(descr1, conf2, sig, rc, r0, alpha_1_descr): """ Eam kernel for global energy-force correlation Args: descr1 (float): descriptor calculated for the first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[2] alpha_1_descr (int): element of the central atom Returns: kernel (array): 3x1 energy-force Eam kernel """ return k_ee_fun_d(np.zeros(3), descr1, conf2, sig, rc, r0, alpha_1_descr) def k2_ef_d(descr1, conf2, sig, rc, r0, alpha_1_descr): """ Eam kernel for force-force correlation Args: descr1 (float): descriptor calculated for the first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] rc (float): cutoff distance hyperparameter theta[1] alpha_1_descr (int): element of the central atom Returns: kernel (matrix): 3x3 force-force Eam kernel """ return -k_ef_fun_d(np.zeros(3), descr1, conf2, sig, rc, r0, alpha_1_descr) logger.info("Ended compilation of theano eam multi species kernels") return k2_ee, k2_ef, k2_ff, k2_ee_d, k2_ef_d
42,667
40.7087
134
py
mff
mff-master/mff/kernels/__init__.py
from .eamkernel import EamManySpeciesKernel, EamSingleSpeciesKernel from .manybodykernel import (ManyBodyManySpeciesKernel, ManyBodySingleSpeciesKernel) from .threebodykernel import (ThreeBodyManySpeciesKernel, ThreeBodySingleSpeciesKernel) from .twobodykernel import TwoBodyManySpeciesKernel, TwoBodySingleSpeciesKernel __all__ = [TwoBodySingleSpeciesKernel, TwoBodyManySpeciesKernel, ThreeBodySingleSpeciesKernel, ThreeBodyManySpeciesKernel, ManyBodySingleSpeciesKernel, ManyBodyManySpeciesKernel, EamSingleSpeciesKernel, EamManySpeciesKernel]
684
41.8125
79
py
mff
mff-master/mff/kernels/manybodykernel.py
# -*- coding: utf-8 -*- import logging import os.path import pickle from abc import ABCMeta, abstractmethod import numpy as np from mff.kernels.base import Kernel, Mffpath logger = logging.getLogger(__name__) def dummy_calc_ff(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype = data if kertype == "single": with open(Mffpath / "k3_ff_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "k3_ff_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros((len(array), 3, 3)) for i in np.arange(len(array)): result[i] = fun(np.zeros(3), np.zeros(3), array[i][0], array[i][1], theta0, theta1, theta2) return result def dummy_calc_ee(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel valuesf Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype = data if kertype == "single": with open(Mffpath / "k3_ee_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "k3_ee_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros(len(array)) for i in np.arange(len(array)): for conf1 in array[i][0]: for conf2 in array[i][1]: result[i] += fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) return result def dummy_calc_ef(data): """ Function used when multiprocessing. Args: data (list of objects): contains all the information required for the computation of the kernel values Returns: result (array): the computed kernel values """ array, theta0, theta1, theta2, kertype = data if kertype == "single": with open(Mffpath / "k3_ef_s.pickle", 'rb') as f: fun = pickle.load(f) elif kertype == "multi": with open(Mffpath / "k3_ef_m.pickle", 'rb') as f: fun = pickle.load(f) result = np.zeros((len(array), 3)) for i in np.arange(len(array)): conf2 = np.array(array[i][1], dtype='float') for conf1 in array[i][0]: conf1 = np.array(conf1, dtype='float') result[i] += -fun(np.zeros(3), np.zeros(3), conf1, conf2, theta0, theta1, theta2) return result class BaseManyBody(Kernel, metaclass=ABCMeta): """ Many body kernel class Handles the functions common to the single-species and multi-species three-body kernels. Args: kernel_name (str): To choose between single- and two-species kernel theta[0] (float) : lengthscale of the kernel theta[1] (float) : decay rate of the cutoff function theta[2] (float) : cutoff radius bounds (list) : bounds of the kernel function. Attributes: km_ee (object): Energy-energy kernel function km_ef (object): Energy-force kernel function km_ff (object): Force-force kernel function """ @abstractmethod def __init__(self, kernel_name, theta, bounds): super().__init__(kernel_name) self.theta = theta self.bounds = bounds self.km_ee, self.km_ef, self.km_ff = self.compile_theano() def calc(self, X1, X2, ncores=1): """ Calculate the energy-force kernel between two sets of configurations. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N2*3 matrix of the vector-valued kernels """ ker = np.zeros((len(X1) * 3, len(X2) * 3)) if ncores > 1: confs = [] for x1 in X1: for x2 in X2: confs.append(np.asarray([x1, x2])) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 3-body force-force kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ff, clist) pool.close() pool.join() result = np.concatenate(result).reshape((n, 3, 3)) for i in range(len(X1)): for j in range(len(X2)): ker[i * 3: i * 3 + 3, 3 * j:3 * j + 3] = result[(j + i * len(X2))] else: for i, conf1 in enumerate(X1): for j, conf2 in enumerate(X2): ker[i * 3:i * 3 + 3, 3 * j:3 * j + 3] += self.km_ff( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_ef(self, X_glob, X, ncores=1, mapping = False): """ Calculate the energy-force kernel between two sets of configurations. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N2*3 matrix of the vector-valued kernels """ ker = np.zeros((len(X_glob), len(X) * 3)) if ncores > 1: confs = [] for x1 in X_glob: for x2 in X: confs.append(np.asarray([x1, x2])) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 3-body energy-force kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ef, clist) pool.close() pool.join() result = np.vstack(np.asarray(result)) for i in range(len(X_glob)): for j in range(len(X)): ker[i, 3 * j:3 * j + 3] = result[(j + i * len(X))] else: for i, x1 in enumerate(X_glob): for j, conf2 in enumerate(X): for conf1 in x1: ker[i, 3 * j:3 * j + 3] += self.km_ef( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_ee(self, X1, X2, ncores=1, mapping = False): """ Calculate the energy-energy kernel between two global environments. Args: X1 (list): list of N1 Mx5 arrays containing xyz coordinates and atomic species X2 (list): list of N2 Mx5 arrays containing xyz coordinates and atomic species Returns: K (matrix): N1 x N2 matrix of the scalar-valued kernels """ if ncores > 1: # Used for multiprocessing confs = [] # Build a list of all input pairs which matrix needs to be computed for x1 in X1: for x2 in X2: confs.append(np.asarray([x1, x2])) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the 3-body energy-energy kernel calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ee, clist) pool.close() pool.join() result = np.concatenate(result).ravel() ker = np.zeros((len(X1), len(X2))) for i in range(len(X1)): for j in range(len(X2)): ker[i, j] = result[j + i*len(X2)] else: ker = np.zeros((len(X1), len(X2))) for i, x1 in enumerate(X1): for j, x2 in enumerate(X2): for conf1 in x1: for conf2 in x2: ker[i, j] += self.km_ee(conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) return ker def calc_gram(self, X, ncores=1, eval_gradient=False): """ Calculate the force-force gram matrix for a set of configurations X. Args: X (list): list of N Mx5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N*3 x N*3 gram matrix of the matrix-valued kernels """ if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: confs = [] for i in np.arange(len(X)): for j in np.arange(i + 1): thislist = np.asarray([X[i], X[j]]) confs.append(thislist) n = len(confs) logger.info( 'Using %i cores for the many-body force-force gram matrix calculation' % (ncores)) import sys sys.setrecursionlimit(100000) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ff, clist) pool.close() pool.join() result = np.concatenate(result).reshape((n, 3, 3)) off_diag = np.zeros((len(X) * 3, len(X) * 3)) diag = np.zeros((len(X) * 3, len(X) * 3)) for i in np.arange(len(X)): diag[3 * i:3 * i + 3, 3 * i:3 * i + 3] = result[i + i * (i + 1) // 2] for j in np.arange(i): off_diag[3 * i:3 * i + 3, 3 * j:3 * j + 3] = result[j + i * (i + 1) // 2] else: diag = np.zeros((X.shape[0] * 3, X.shape[0] * 3)) off_diag = np.zeros((X.shape[0] * 3, X.shape[0] * 3)) for i in np.arange(X.shape[0]): diag[3 * i:3 * i + 3, 3 * i:3 * i + 3] = \ self.km_ff(X[i], X[i], self.theta[0], self.theta[1], self.theta[2]) for j in np.arange(i): off_diag[3 * i:3 * i + 3, 3 * j:3 * j + 3] = \ self.km_ff(X[i], X[j], self.theta[0], self.theta[1], self.theta[2]) gram = diag + off_diag + off_diag.T return gram def calc_gram_e(self, X, ncores=1, eval_gradient=False): # Untested """ Calculate the energy-energy gram matrix for a set of configurations X. Args: X (list): list of N Mx5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N x N gram matrix of the scalar-valued kernels """ if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: confs = [] # Build a list of all input pairs which matrix needs to be computed for i in np.arange(len(X)): for j in np.arange(i + 1): thislist = np.array([list(X[i]), list(X[j])]) confs.append(thislist) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the many-body energy-energy gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ee, clist) pool.close() pool.join() result = np.concatenate(result).ravel() off_diag = np.zeros((len(X), len(X))) diag = np.zeros((len(X), len(X))) for i in np.arange(len(X)): diag[i, i] = result[i + i * (i + 1) // 2] for j in np.arange(i): off_diag[i, j] = result[j + i * (i + 1) // 2] else: diag = np.zeros((X.shape[0], X.shape[0])) off_diag = np.zeros((X.shape[0], X.shape[0])) for i in np.arange(X.shape[0]): for k, conf1 in enumerate(X[i]): diag[i, i] += self.km_ee(conf1, conf1, self.theta[0], self.theta[1], self.theta[2]) for conf2 in X[i][:k]: # *2 here to speed up the loop diag[i, i] += 2.0*self.km_ee( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) for j in np.arange(i): for conf1 in X[i]: for conf2 in X[j]: off_diag[i, j] += self.km_ee( conf1, conf2, self.theta[0], self.theta[1], self.theta[2]) gram = diag + off_diag + off_diag.T # Gram matrix is symmetric return gram def calc_gram_ef(self, X, X_glob, ncores=1, eval_gradient=False): """ Calculate the energy-force gram matrix for a set of configurations X. This returns a non-symmetric matrix which is equal to the transpose of the force-energy gram matrix. Args: X (list): list of N1 M1x5 arrays containing xyz coordinates and atomic species X_glob (list): list of N2 M2x5 arrays containing xyz coordinates and atomic species ncores (int): Number of CPU nodes to use for multiprocessing (default is 1) eval_gradient (bool): if True, evaluate the gradient of the gram matrix Returns: gram (matrix): N2 x N1*3 gram matrix of the vector-valued kernels """ gram = np.zeros((X_glob.shape[0], X.shape[0] * 3)) if eval_gradient: raise NotImplementedError('ERROR: GRADIENT NOT IMPLEMENTED YET') else: if ncores > 1: # Multiprocessing confs = [] for i in np.arange(len(X_glob)): for j in np.arange(len(X)): thislist = np.asarray([X_glob[i], X[j]]) confs.append(thislist) n = len(confs) import sys sys.setrecursionlimit(100000) logger.info( 'Using %i cores for the many-body energy-force gram matrix calculation' % (ncores)) # Way to split the kernels functions to compute evenly across the nodes splitind = np.zeros(ncores + 1) factor = (n + (ncores - 1)) / ncores splitind[1:-1] = [(i + 1) * factor for i in np.arange(ncores - 1)] splitind[-1] = n splitind = splitind.astype(int) clist = [[confs[splitind[i]:splitind[i + 1]], self.theta[0], self.theta[1], self.theta[2], self.type] for i in np.arange(ncores)] # Shape is ncores * (ntrain*(ntrain+1)/2)/ncores import multiprocessing as mp pool = mp.Pool(ncores) result = pool.map(dummy_calc_ef, clist) pool.close() pool.join() result = np.concatenate(result).ravel() for i in np.arange(X_glob.shape[0]): for j in np.arange(X.shape[0]): gram[i, 3 * j:3 * j + 3] = result[3 * (j + i * X.shape[0]):3 + 3*(j + i * X.shape[0])] else: for i in np.arange(X_glob.shape[0]): for j in np.arange(X.shape[0]): for k in X_glob[i]: gram[i, 3 * j:3 * j + 3] += self.km_ef( k, X[j], self.theta[0], self.theta[1], self.theta[2]) self.gram_ef = gram return gram def calc_diag(self, X): diag = np.zeros((X.shape[0] * 3)) for i in np.arange(X.shape[0]): diag[i * 3:(i + 1) * 3] = np.diag(self.km_ff(X[i], X[i], self.theta[0], self.theta[1], self.theta[2])) return diag def calc_diag_e(self, X): diag = np.zeros((X.shape[0])) for i in np.arange(X.shape[0]): diag[i] = self.km_ee(X[i], X[i], self.theta[0], self.theta[1], self.theta[2]) return diag @staticmethod @abstractmethod def compile_theano(): return None, None, None class ManyBodySingleSpeciesKernel(BaseManyBody): """Many body two species kernel. Args: theta[0] (float): lengthscale of the kernel theta[1] (float): decay rate of the cutoff function theta[2] (float): cutoff radius """ def __init__(self, theta=(1., 1., 1.), bounds=((1e-2, 1e2), (1e-2, 1e2), (1e-2, 1e2))): super().__init__(kernel_name='ManyBodySingleSpecies', theta=theta, bounds=bounds) self.type = "single" @staticmethod def compile_theano(): """ This function generates theano compiled kernels for energy and force learning ker_jkmn_withcutoff = ker_jkmn #* cutoff_ikmn The position of the atoms relative to the centrla one, and their chemical species are defined by a matrix of dimension Mx5 Returns: km_ee (func): energy-energy kernel km_ef (func): energy-force kernel km_ff (func): force-force kernel """ if not (os.path.exists(Mffpath / 'k3_ee_s.pickle') and os.path.exists(Mffpath / 'k3_ef_s.pickle') and os.path.exists(Mffpath / 'k3_ff_s.pickle')): print("Building Kernels") import theano.tensor as T from theano import function, scan logger.info("Started compilation of theano three body kernels") # -------------------------------------------------- # INITIAL DEFINITIONS # -------------------------------------------------- # positions of central atoms r1, r2 = T.dvectors('r1d', 'r2d') # positions of neighbours rho1, rho2 = T.dmatrices('rho1', 'rho2') # hyperparameter sig = T.dscalar('sig') # cutoff hyperparameters theta = T.dscalar('theta') rc = T.dscalar('rc') # positions of neighbours without chemical species rho1s = rho1[:, 0:3] rho2s = rho2[:, 0:3] # -------------------------------------------------- # RELATIVE DISTANCES TO CENTRAL VECTOR AND BETWEEN NEIGHBOURS # -------------------------------------------------- # first and second configuration r1j = T.sqrt(T.sum((rho1s[:, :] - r1[None, :]) ** 2, axis=1)) r2m = T.sqrt(T.sum((rho2s[:, :] - r2[None, :]) ** 2, axis=1)) rjk = T.sqrt( T.sum((rho1s[None, :, :] - rho1s[:, None, :]) ** 2, axis=2)) rmn = T.sqrt( T.sum((rho2s[None, :, :] - rho2s[:, None, :]) ** 2, axis=2)) # -------------------------------------------------- # BUILD THE KERNEL # -------------------------------------------------- # Squared exp of differences se_1j2m = T.exp(-(r1j[:, None] - r2m[None, :]) ** 2 / (2 * sig ** 2)) se_jkmn = T.exp(-(rjk[:, :, None, None] - rmn[None, None, :, :]) ** 2 / (2 * sig ** 2)) se_jk2m = T.exp(-(rjk[:, :, None] - r2m[None, None, :]) ** 2 / (2 * sig ** 2)) se_1jmn = T.exp(-(r1j[:, None, None] - rmn[None, :, :]) ** 2 / (2 * sig ** 2)) # Kernel not summed (cyclic permutations) k1n = (se_1j2m[:, None, :, None] * se_1j2m[None, :, None, :] * se_jkmn) k2n = (se_1jmn[:, None, :, :] * se_jk2m[:, :, None, :] * se_1j2m[None, :, :, None]) k3n = (se_1j2m[:, None, None, :] * se_jk2m[:, :, :, None] * se_1jmn[None, :, :, :]) # final shape is M1 M1 M2 M2 ker = k1n + k2n + k3n cut_j = 0.5*(1+T.cos(np.pi*r1j/rc)) cut_m = 0.5*(1+T.cos(np.pi*r2m/rc)) cut_jk = cut_j[:,None]*cut_j[None,:]*0.5*(1+T.cos(np.pi*rjk/rc)) cut_mn = cut_m[:,None]*cut_m[None,:]*0.5*(1+T.cos(np.pi*rmn/rc)) # -------------------------------------------------- # REMOVE DIAGONAL ELEMENTS AND ADD CUTOFF # -------------------------------------------------- # remove diagonal elements AND lower triangular ones from first configuration mask_jk = T.triu(T.ones_like(rjk)) - T.identity_like(rjk) # remove diagonal elements from second configuration mask_mn = T.ones_like(rmn) - T.identity_like(rmn) # Combine masks mask_jkmn = mask_jk[:, :, None, None] * mask_mn[None, None, :, :] # Apply mask and then apply cutoff functions ker = ker * mask_jkmn ker = T.sum(ker * cut_jk[:, :, None, None] * cut_mn[None, None, :, :]) ker = T.exp(ker / 1000) # -------------------------------------------------- # FINAL FUNCTIONS # -------------------------------------------------- # global energy energy kernel k_ee_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], ker, on_unused_input='ignore') # global energy force kernel k_ef = T.grad(ker, r2) k_ef_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], k_ef, on_unused_input='ignore') # local force force kernel k_ff = T.grad(ker, r1) k_ff_der, updates = scan(lambda j, k_ff, r2: T.grad(k_ff[j], r2), sequences=T.arange(k_ff.shape[0]), non_sequences=[k_ff, r2]) k_ff_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], k_ff_der, on_unused_input='ignore') # Save the function that we want to use for multiprocessing # This is necessary because theano is a crybaby and does not want to access the # Automaticallly stored compiled object from different processes with open(Mffpath / 'k3_ee_s.pickle', 'wb') as f: pickle.dump(k_ee_fun, f) with open(Mffpath / 'k3_ef_s.pickle', 'wb') as f: pickle.dump(k_ef_fun, f) with open(Mffpath / 'k3_ff_s.pickle', 'wb') as f: pickle.dump(k_ff_fun, f) else: print("Loading Kernels") with open(Mffpath / "k3_ee_s.pickle", 'rb') as f: k_ee_fun = pickle.load(f) with open(Mffpath / "k3_ef_s.pickle", 'rb') as f: k_ef_fun = pickle.load(f) with open(Mffpath / "k3_ff_s.pickle", 'rb') as f: k_ff_fun = pickle.load(f) # WRAPPERS (we don't want to plug the position of the central element every time) def km_ee(conf1, conf2, sig, theta, rc): """ Many body kernel for global energy-energy correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (float): scalar valued energy-energy many-body kernel """ return k_ee_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def km_ef(conf1, conf2, sig, theta, rc): """ Many body kernel for global energy-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (array): 3x1 energy-force many-body kernel """ return -k_ef_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def km_ff(conf1, conf2, sig, theta, rc): """ Many body kernel for local force-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (matrix): 3x3 force-force 3-body kernel """ return k_ff_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) logger.info("Ended compilation of theano three body kernels") return km_ee, km_ef, km_ff class ManyBodyManySpeciesKernel(BaseManyBody): """Many body many species kernel. Args: theta[0] (float): lengthscale of the kernel theta[1] (float): decay rate of the cutoff function theta[2] (float): cutoff radius """ def __init__(self, theta=(1., 1., 1.), bounds=((1e-2, 1e2), (1e-2, 1e2), (1e-2, 1e2))): super().__init__(kernel_name='ManyBodyManySpecies', theta=theta, bounds=bounds) self.type = "multi" @staticmethod def compile_theano(): """ This function generates theano compiled kernels for energy and force learning ker_jkmn_withcutoff = ker_jkmn #* cutoff_ikmn The position of the atoms relative to the centrla one, and their chemical species are defined by a matrix of dimension Mx5 Returns: km_ee (func): energy-energy kernel km_ef (func): energy-force kernel km_ff (func): force-force kernel """ if not (os.path.exists(Mffpath / 'k3_ee_m.pickle') and os.path.exists(Mffpath / 'k3_ef_m.pickle') and os.path.exists(Mffpath / 'k3_ff_m.pickle')): print("Building Kernels") import theano.tensor as T from theano import function, scan logger.info("Started compilation of theano three body kernels") # -------------------------------------------------- # INITIAL DEFINITIONS # -------------------------------------------------- # positions of central atoms r1, r2 = T.dvectors('r1d', 'r2d') # positions of neighbours rho1, rho2 = T.dmatrices('rho1', 'rho2') # hyperparameter sig = T.dscalar('sig') # cutoff hyperparameters theta = T.dscalar('theta') rc = T.dscalar('rc') # positions of neighbours without chemical species rho1s = rho1[:, 0:3] rho2s = rho2[:, 0:3] alpha_1 = rho1[:, 3].flatten() alpha_2 = rho2[:, 3].flatten() alpha_j = rho1[:, 4].flatten() alpha_m = rho2[:, 4].flatten() alpha_k = rho1[:, 4].flatten() alpha_n = rho2[:, 4].flatten() # -------------------------------------------------- # RELATIVE DISTANCES TO CENTRAL VECTOR AND BETWEEN NEIGHBOURS # -------------------------------------------------- # first and second configuration r1j = T.sqrt(T.sum((rho1s[:, :] - r1[None, :]) ** 2, axis=1)) r2m = T.sqrt(T.sum((rho2s[:, :] - r2[None, :]) ** 2, axis=1)) rjk = T.sqrt( T.sum((rho1s[None, :, :] - rho1s[:, None, :]) ** 2, axis=2)) rmn = T.sqrt( T.sum((rho2s[None, :, :] - rho2s[:, None, :]) ** 2, axis=2)) # -------------------------------------------------- # CHEMICAL SPECIES MASK # -------------------------------------------------- # numerical kronecker def delta_alpha2(a1j, a2m): d = np.exp(-(a1j - a2m) ** 2 / (2 * 0.00001 ** 2)) return d # permutation 1 delta_alphas12 = delta_alpha2(alpha_1[0], alpha_2[0]) delta_alphasjm = delta_alpha2(alpha_j[:, None], alpha_m[None, :]) delta_alphas_jmkn = delta_alphasjm[:, None, :, None] * delta_alphasjm[None, :, None, :] delta_perm1 = delta_alphas12 * delta_alphas_jmkn # permutation 3 delta_alphas1m = delta_alpha2( alpha_1[0, None], alpha_m[None, :]).flatten() delta_alphasjn = delta_alpha2(alpha_j[:, None], alpha_n[None, :]) delta_alphask2 = delta_alpha2( alpha_k[:, None], alpha_2[None, 0]).flatten() delta_perm3 = delta_alphas1m[None, None, :, None] * delta_alphasjn[:, None, None, :] * \ delta_alphask2[None, :, None, None] # permutation 5 delta_alphas1n = delta_alpha2( alpha_1[0, None], alpha_n[None, :]).flatten() delta_alphasj2 = delta_alpha2( alpha_j[:, None], alpha_2[None, 0]).flatten() delta_alphaskm = delta_alpha2(alpha_k[:, None], alpha_m[None, :]) delta_perm5 = delta_alphas1n[None, None, None, :] * delta_alphaskm[None, :, :, None] * \ delta_alphasj2[:, None, None, None] # -------------------------------------------------- # BUILD THE KERNEL # -------------------------------------------------- # Squared exp of differences se_1j2m = T.exp(-(r1j[:, None] - r2m[None, :]) ** 2 / (2 * sig ** 2)) se_jkmn = T.exp(-(rjk[:, :, None, None] - rmn[None, None, :, :]) ** 2 / (2 * sig ** 2)) se_jk2m = T.exp(-(rjk[:, :, None] - r2m[None, None, :]) ** 2 / (2 * sig ** 2)) se_1jmn = T.exp(-(r1j[:, None, None] - rmn[None, :, :]) ** 2 / (2 * sig ** 2)) # Kernel not summed (cyclic permutations) k1n = (se_1j2m[:, None, :, None] * se_1j2m[None, :, None, :] * se_jkmn) k2n = (se_1jmn[:, None, :, :] * se_jk2m[:, :, None, :] * se_1j2m[None, :, :, None]) k3n = (se_1j2m[:, None, None, :] * se_jk2m[:, :, :, None] * se_1jmn[None, :, :, :]) # final shape is M1 M1 M2 M2 ker_loc = k1n * delta_perm1 + k2n * delta_perm3 + k3n * delta_perm5 # Faster version of cutoff (less calculations) cut_j = 0.5*(1+T.cos(np.pi*r1j/rc)) cut_m = 0.5*(1+T.cos(np.pi*r2m/rc)) cut_jk = cut_j[:,None]*cut_j[None,:]*0.5*(1+T.cos(np.pi*rjk/rc)) cut_mn = cut_m[:,None]*cut_m[None,:]*0.5*(1+T.cos(np.pi*rmn/rc)) # -------------------------------------------------- # REMOVE DIAGONAL ELEMENTS # -------------------------------------------------- # remove diagonal elements AND lower triangular ones from first configuration mask_jk = T.triu(T.ones_like(rjk)) - T.identity_like(rjk) # remove diagonal elements from second configuration mask_mn = T.ones_like(rmn) - T.identity_like(rmn) # Combine masks mask_jkmn = mask_jk[:, :, None, None] * mask_mn[None, None, :, :] # Apply mask and then apply cutoff functions ker_loc = ker_loc * mask_jkmn ker_loc = T.sum( ker_loc * cut_jk[:, :, None, None] * cut_mn[None, None, :, :]) ker_loc = T.exp(ker_loc / 20) # -------------------------------------------------- # FINAL FUNCTIONS # -------------------------------------------------- # energy energy kernel k_ee_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], ker_loc, on_unused_input='ignore') # energy force kernel k_ef_cut = T.grad(ker_loc, r2) k_ef_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], k_ef_cut, on_unused_input='ignore') # force force kernel k_ff_cut = T.grad(ker_loc, r1) k_ff_cut_der, updates = scan(lambda j, k_ff_cut, r2: T.grad(k_ff_cut[j], r2), sequences=T.arange(k_ff_cut.shape[0]), non_sequences=[k_ff_cut, r2]) k_ff_fun = function( [r1, r2, rho1, rho2, sig, theta, rc], k_ff_cut_der, on_unused_input='ignore') # Save the function that we want to use for multiprocessing # This is necessary because theano is a crybaby and does not want to access the # Automaticallly stored compiled object from different processes with open(Mffpath / 'k3_ee_m.pickle', 'wb') as f: pickle.dump(k_ee_fun, f) with open(Mffpath / 'k3_ef_m.pickle', 'wb') as f: pickle.dump(k_ef_fun, f) with open(Mffpath / 'k3_ff_m.pickle', 'wb') as f: pickle.dump(k_ff_fun, f) else: print("Loading Kernels") with open(Mffpath / "k3_ee_m.pickle", 'rb') as f: k_ee_fun = pickle.load(f) with open(Mffpath / "k3_ef_m.pickle", 'rb') as f: k_ef_fun = pickle.load(f) with open(Mffpath / "k3_ff_m.pickle", 'rb') as f: k_ff_fun = pickle.load(f) # WRAPPERS (we don't want to plug the position of the central element every time) def km_ee(conf1, conf2, sig, theta, rc): """ Many body kernel for energy-energy correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (float): scalar valued energy-energy many-body kernel """ return k_ee_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def km_ef(conf1, conf2, sig, theta, rc): """ Many body kernel for energy-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (array): 3x1 energy-force many-body kernel """ return -k_ef_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) def km_ff(conf1, conf2, sig, theta, rc): """ Many body kernel for force-force correlation Args: conf1 (array): first configuration. conf2 (array): second configuration. sig (float): lengthscale hyperparameter theta[0] theta (float): cutoff decay rate hyperparameter theta[1] rc (float): cutoff distance hyperparameter theta[2] Returns: kernel (matrix): 3x3 force-force many-body kernel """ return k_ff_fun(np.zeros(3), np.zeros(3), conf1, conf2, sig, theta, rc) logger.info("Ended compilation of theano many body kernels") return km_ee, km_ef, km_ff
40,153
38.994024
114
py
T-Concord3D
T-Concord3D-master/test.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------| import os import argparse import sys import matplotlib.pyplot as plt import numpy as np import torch import torch.optim as optim from tqdm import tqdm import math from utils.metric_util import per_class_iu, fast_hist_crop, fast_ups_crop from dataloader.pc_dataset import get_label_name, get_label_inv_name, update_config from builder import data_builder, model_builder, loss_builder from config.config import load_config_data import torch.nn.functional as F from utils.load_save_util import load_checkpoint from utils.ups import enable_dropout import warnings from torch.nn.parallel import DistributedDataParallel warnings.filterwarnings("ignore") def save_predictions_sematicKitti(predict_labels_serialized, predict_prob_serialized, path_to_seq_folder, path_to_seq_folder_prob, sample_name, challenge=False): # dump predictions and probability predict_labels_serialized.tofile(path_to_seq_folder + '/' + sample_name + '.label') if not challenge: if not os.path.exists(path_to_seq_folder_prob): os.makedirs(path_to_seq_folder_prob) predict_prob_serialized.tofile(path_to_seq_folder_prob + '/' + sample_name + '.label') def save_predictions_wod(predict_labels_serialized, predict_prob_serialized, path_to_seq_folder, path_to_seq_folder_prob, sample_name, challenge=False): # dump predictions and probability np.save(os.path.join(path_to_seq_folder, sample_name), predict_labels_serialized) if not challenge: if not os.path.exists(path_to_seq_folder_prob): os.makedirs(path_to_seq_folder_prob) np.save(os.path.join(path_to_seq_folder_prob, sample_name), predict_prob_serialized) def main(args): os.environ['OMP_NUM_THREADS'] = "1" distributed = False if "WORLD_SIZE" in os.environ: distributed = int(os.environ["WORLD_SIZE"]) > 1 print(f"distributed: {distributed}") pytorch_device = args.local_rank if distributed: torch.cuda.set_device(pytorch_device) torch.distributed.init_process_group(backend='nccl', init_method='env://') args.world_size = torch.distributed.get_world_size() config_path = args.config_path configs = load_config_data(config_path) if args.mode == 'infer' or args.mode == 'val' or args.mode == 'test': configs['train_params']['ssl'] = False # send config parameters to pc_dataset update_config(configs) dataset_config = configs['dataset_params'] dataset_type = 'SemanticKITTI' if 'SemKITTI_sk_multiscan' == dataset_config['pc_dataset_type'] else 'WOD' train_dataloader_config = configs['train_data_loader'] ssl_dataloader_config = configs['ssl_data_loader'] val_dataloader_config = configs['val_data_loader'] test_dataloader_config = configs['test_data_loader'] val_batch_size = val_dataloader_config['batch_size'] train_batch_size = train_dataloader_config['batch_size'] ssl_batch_size = ssl_dataloader_config['batch_size'] test_batch_size = test_dataloader_config['batch_size'] model_config = configs['model_params'] train_hypers = configs['train_params'] past_frame = train_hypers['past'] future_frame = train_hypers['future'] T_past_frame = train_hypers['T_past'] T_future_frame = train_hypers['T_future'] grid_size = model_config['output_shape'] num_class = model_config['num_class'] ignore_label = dataset_config['ignore_label'] model_load_path = train_hypers['model_load_path'] model_save_path = train_hypers['model_save_path'] SemKITTI_label_name = get_label_name(dataset_config["label_mapping"]) unique_label = np.asarray(sorted(list(SemKITTI_label_name.keys())))[1:] - 1 unique_label_str = [SemKITTI_label_name[x] for x in unique_label + 1] print(unique_label_str) SemKITTI_learningmap_inv = get_label_inv_name(dataset_config["label_mapping"]) model = model_builder.build(model_config).to(pytorch_device) print(f"model_load_path: {model_load_path}") if os.path.exists(model_load_path): model = load_checkpoint(model_load_path, model, map_location=pytorch_device) print(f" loading model_load_path: {model_load_path}") # if args.mgpus: # my_model = nn.DataParallel(my_model) # #my_model.cuda() # #my_model.cuda() if distributed: model = DistributedDataParallel( model, device_ids=[pytorch_device], output_device=args.local_rank, find_unused_parameters=True ) optimizer = optim.Adam(model.parameters(), lr=train_hypers["learning_rate"]) loss_func, lovasz_softmax = loss_builder.build(wce=True, lovasz=True, num_class=num_class, ignore_label=ignore_label) train_dataset_loader, val_dataset_loader, test_dataset_loader, ssl_dataset_loader = data_builder.build( dataset_config, train_dataloader_config, val_dataloader_config, test_dataloader_config, ssl_dataloader_config, grid_size=grid_size, train_hypers=train_hypers) # test and validation if args.mode == 'val': dataset_loader = val_dataset_loader batch_size = val_batch_size path_to_save_predicted_labels = val_dataloader_config['data_path'] # "val_result" elif args.mode == 'test': dataset_loader = test_dataset_loader batch_size = test_batch_size path_to_save_predicted_labels = test_dataloader_config['data_path'] # "test_result" elif args.mode == 'infer': dataset_loader = ssl_dataset_loader batch_size = ssl_batch_size path_to_save_predicted_labels = ssl_dataloader_config['data_path'] # "pseudo_label_result" # mode to eval model.eval() # if uncertainty is used, enable dropout if args.ups: # enable dropout (mc) enable_dropout(model) # sample forward pass f_pass = 10 with torch.no_grad(): ups_hist = [] hist_list = [] hist_list_op = [] ups_count = [] def validation_inference(vox_label, grid, pt_labs, pt_fea, ref_st_idx=None, ref_end_idx=None, lcw=None): val_pt_fea_ten = [torch.from_numpy(i).type(torch.FloatTensor).to(pytorch_device) for i in pt_fea] val_grid_ten = [torch.from_numpy(i).to(pytorch_device) for i in grid] if args.ups: ups_out_prob = [] for _ in range(f_pass): predict_labels_raw = model(val_pt_fea_ten, val_grid_ten, batch_size) ups_out_prob.append(F.softmax(predict_labels_raw, dim=1)) # for selecting positive pseudo-labels ups_out_prob = torch.stack(ups_out_prob) out_std = torch.std(ups_out_prob, dim=0) predict_probablity = torch.mean(ups_out_prob, dim=0) predict_labels = torch.argmax(predict_probablity, dim=1) # keep dimension during finding maximum predict_prob_max, predict_prob_ind = torch.max(predict_probablity, dim=1, keepdim=True) # squeeze (remove the 1 size form the tensor) predict_prob_max = torch.squeeze(predict_prob_max) # get the uncertainty of the most probable prediction max_std = out_std.gather(1, predict_prob_ind) # squeeze (remove the 1 size form the tensor) max_std = torch.squeeze(max_std) else: predict_labels_raw = model(val_pt_fea_ten, val_grid_ten, batch_size) predict_labels = torch.argmax(predict_labels_raw, dim=1) predict_probablity = torch.nn.functional.softmax(predict_labels_raw, dim=1) predict_prob_max, predict_prob_ind = predict_probablity.max(dim=1) # move to cpu and detach to convert to numpy predict_labels = predict_labels.cpu().detach().numpy() predict_probabilitys = predict_prob_max.cpu().detach().numpy() if args.ups: model_uncertintys = max_std.cpu().detach().numpy() for count, i_val_grid in enumerate(grid): if args.save_raw: predict_raw = predict_labels_raw[count, grid[count][:, 0], grid[count][:, 1], grid[count][:, 2]] predict_label = predict_labels[count, grid[count][:, 0], grid[count][:, 1], grid[count][:, 2]] predict_prob = predict_probabilitys[count, grid[count][:, 0], grid[count][:, 1], grid[count][:, 2]] if args.ups: model_uncertainty = model_uncertintys[ count, grid[count][:, 0], grid[count][:, 1], grid[count][:, 2]] model_uncertainty_serialized = np.array(model_uncertainty, dtype=np.float32) predict_labels_serialized = np.array(predict_label, dtype=np.int32) predict_prob_serialized = np.array(predict_prob, dtype=np.float32) if args.save_raw: predict_raw_serialized = np.array(predict_raw, dtype=np.float32) demo_pt_labs = pt_labs[count] # get reference frame start and end index st_id, end_id = int(ref_st_idx[count]), int(ref_end_idx[count]) # only select the reference frame points if ref_st_idx is not None: predict_labels_serialized = predict_labels_serialized[st_id:st_id + end_id] predict_prob_serialized = predict_prob_serialized[st_id:st_id + end_id] if args.save_raw: predict_raw_serialized = predict_raw_serialized[st_id:st_id + end_id] demo_pt_labs = demo_pt_labs[st_id:st_id + end_id] if args.ups: model_uncertainty_serialized = model_uncertainty_serialized[st_id:st_id + end_id] if args.mode == 'val': hist_list.append(fast_hist_crop(predict_labels_serialized, demo_pt_labs, unique_label)) if args.ups: tmp_hist, temp_count = fast_ups_crop(model_uncertainty_serialized, demo_pt_labs.flatten(), unique_label) ups_hist.append(tmp_hist) ups_count.append(temp_count) if args.save: # convert the prediction into corresponding GT labels (inverse mapping) # for index, label in enumerate(predict_labels_serialized): # predict_labels_serialized[index] = SemKITTI_learningmap_inv[label] # print(predict_labels_serialized.size) predict_labels_serialized = np.vectorize(SemKITTI_learningmap_inv.__getitem__)(predict_labels_serialized) # get frame and sequence name sample_name = dataset_loader.dataset.point_cloud_dataset.im_idx[i_iter_val * batch_size + count][ -10:-4] sequence_num = dataset_loader.dataset.point_cloud_dataset.im_idx[i_iter_val * batch_size + count].split('/')[-3] # create destination path to save predictions # path_to_seq_folder = path_to_save_predicted_labels + '/' + str(sequence_num) path_to_seq_folder = os.path.join(path_to_save_predicted_labels, str(sequence_num), f"predictions_f{T_past_frame}_{T_future_frame}") path_to_seq_folder_prob = os.path.join(path_to_save_predicted_labels, str(sequence_num), f"probability_f{T_past_frame}_{T_future_frame}") if args.save_raw: path_to_seq_folder_raw = os.path.join(path_to_save_predicted_labels, str(sequence_num), f"raw_f{T_past_frame}_{T_future_frame}") if args.challenge: path_to_save_test_predicted_labels = args.challenge_path path_to_seq_folder = os.path.join(path_to_save_test_predicted_labels, f"f{T_past_frame}_{T_future_frame}", "sequences", str(sequence_num), "predictions") if not os.path.exists(path_to_seq_folder): os.makedirs(path_to_seq_folder) # dump predictions and probability predict_labels_serialized.tofile(path_to_seq_folder + '/' + sample_name + '.label') if dataset_type == 'SemanticKITTI': save_predictions_sematicKitti(predict_labels_serialized, predict_prob_serialized, path_to_seq_folder, path_to_seq_folder_prob, sample_name, challenge=args.challenge) elif dataset_type == 'WOD': save_predictions_wod(predict_labels_serialized, predict_prob_serialized, path_to_seq_folder, path_to_seq_folder_prob, sample_name, challenge=args.challenge) else: raise Exception(f'{dataset_type} dataset type not known') # if not args.challenge: # if not os.path.exists(path_to_seq_folder_prob): # os.makedirs(path_to_seq_folder_prob) # predict_prob_serialized.tofile(path_to_seq_folder_prob + '/' + sample_name + '.label') # if args.save_raw: # if not os.path.exists(path_to_seq_folder_raw): # os.makedirs(path_to_seq_folder_raw) # # predict_prob_serialized.tofile(path_to_seq_folder_raw + '/' + sample_name + '.label') # Validation with multi-frames and ssl: # if past_frame > 0 and train_hypers['ssl']: for i_iter_val, (_, vox_label, grid, pt_labs, pt_fea, ref_st_idx, ref_end_idx, lcw) in tqdm( enumerate(dataset_loader), total=math.ceil(len(dataset_loader.dataset.point_cloud_dataset.im_idx) / batch_size)): # call the validation and inference with validation_inference(vox_label, grid, pt_labs, pt_fea, ref_st_idx=ref_st_idx, ref_end_idx=ref_end_idx, lcw=lcw) # print the validation per class iou and overall miou if args.mode == 'val': iou = per_class_iu(sum(hist_list)) print('Validation per class iou: ') for class_name, class_iou in zip(unique_label_str, iou): print('%s : %.2f%%' % (class_name, class_iou * 100)) val_miou = np.nanmean(iou) * 100 print('Current val miou is %.3f' % val_miou) if args.ups: uncertainty_hist = np.sum(ups_hist, axis=0) / np.sum(ups_count, axis=0) plt.bar(range(20), uncertainty_hist, width=0.4) plt.show() print(uncertainty_hist) if __name__ == '__main__': # Training settings parser = argparse.ArgumentParser(description='') parser.add_argument('-y', '--config_path', default='config/semantickitti/semantickitti_S0_0_T11_33_ssl_s20_p80.yaml') parser.add_argument('-g', '--mgpus', action='store_true', default=False) parser.add_argument('-m', '--mode', default='val') parser.add_argument('-s', '--save', default=True) parser.add_argument('-c', '--challenge', default=False) parser.add_argument('-p', '--challenge_path', default='/mnt/personal/gebreawe/Datasets/RealWorld/semantic-kitti' '/challenge') parser.add_argument('-u', '--ups', default=False) parser.add_argument("--local_rank", default=0, type=int) parser.add_argument('-r', '--save_raw', default=False) args = parser.parse_args() print(' '.join(sys.argv)) print(args) main(args)
16,515
44.750693
137
py
T-Concord3D
T-Concord3D-master/concordance_kitti.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------| import os import time import argparse import sys import numpy as np import glob import os import shutil import random import math # https://github.com/ctu-vras/T-Concord3D.git def main(args): teacher_1 = args.teacher1 teacher_2 = args.teacher2 teacher_3 = args.teacher3 lamda = args.lamda concordance = args.concordance source = args.source destination = args.destination sequence = ["00", "01", "02", "03", "04", "05", "06", "07", "09", "10"] for i, sq in enumerate(sequence): pred_t1 = sorted(glob.glob(os.path.join(source, sq, f"predictions_{teacher_1}", '*.label'))) pred_t2 = sorted(glob.glob(os.path.join(source, sq, f"predictions_{teacher_2}", '*.label'))) probs_t1 = sorted(glob.glob(os.path.join(source, sq, f"probability_{teacher_1}", '*.label'))) probs_t2 = sorted(glob.glob(os.path.join(source, sq, f"probability_{teacher_2}", '*.label'))) if teacher_3 is not None: pred_t3 = sorted(glob.glob(os.path.join(source, sq, f"predictions_{teacher_3}", '*.label'))) probs_t3 = sorted(glob.glob(os.path.join(source, sq, f"probability_{teacher_3}", '*.label'))) frame_len = len(pred_t1) for frame in range(frame_len): frame_name = str(frame).zfill(6) if teacher_3 is not None: pred = np.array([np.fromfile(pred_t1[frame], dtype=np.int32).reshape((-1, 1)), np.fromfile(pred_t2[frame], dtype=np.int32).reshape((-1, 1)), np.fromfile(pred_t3[frame], dtype=np.int32).reshape((-1, 1))]) prob = np.array([np.fromfile(probs_t1[frame], dtype=np.float32).reshape((-1, 1)), np.fromfile(probs_t2[frame], dtype=np.float32).reshape((-1, 1)), np.fromfile(probs_t3[frame], dtype=np.float32).reshape((-1, 1))]) else: pred = np.array([np.fromfile(pred_t1[frame], dtype=np.int32).reshape((-1, 1)), np.fromfile(pred_t2[frame], dtype=np.int32).reshape((-1, 1))]) prob = np.array([np.fromfile(probs_t1[frame], dtype=np.float32).reshape((-1, 1)), np.fromfile(probs_t2[frame], dtype=np.float32).reshape((-1, 1))]) max_pob = prob.max(axis=0) max_pob_id = prob.argmax(axis=0) best_pred = np.zeros_like(pred[0]) for j in range(len(max_pob)): best_pred[j] = pred[int(max_pob_id[j]), j] weight = np.zeros_like(best_pred) for k in range(3): predicted = pred[k] concord = best_pred == predicted weight += concord.astype(int) best_prob = max_pob new_prob = best_prob + ((weight - 1) * lamda) new_prob = np.minimum(np.ones_like(best_prob), new_prob) new_prob = new_prob.astype(np.float32) if not os.path.exists(os.path.join(destination, sq, f"predictions_{concordance}")): os.makedirs(os.path.join(destination, sq, f"predictions_{concordance}")) os.makedirs(os.path.join(destination, sq, f"probability_{concordance}")) best_pred.tofile(os.path.join(destination, sq, f"predictions_{concordance}", frame_name + '.label')) new_prob.tofile(os.path.join(destination, sq, f"probability_{concordance}", frame_name + '.label')) if __name__ == '__main__': parser = argparse.ArgumentParser() parser.add_argument('-l', '--lamda', default=0.1, type=float) parser.add_argument('-b', '--best', default=True, ) parser.add_argument('-x', '--teacher1', required=False, default="f1_1") parser.add_argument('-y', '--teacher2', required=False, default="f2_2") parser.add_argument('-z', '--teacher3', default="f3_3") parser.add_argument('-c', '--concordance', default="11_33") parser.add_argument('-s', '--source', default='/mnt/beegfs/gpu/argoverse-tracking-all-training/semantic-kitti' '/train_pseudo_20/sequences') parser.add_argument('-d', '--destination', default='/mnt/beegfs/gpu/argoverse-tracking-all-training/semantic' '-kitti/train_pseudo_20/sequences') args = parser.parse_args() main(args)
4,467
44.131313
114
py
T-Concord3D
T-Concord3D-master/train_tconcord3d.py
# -*- coding:utf-8 -*- # author: Awet import argparse import os import sys import time import warnings import numpy as np import torch import torch.optim as optim from torch.nn.parallel import DistributedDataParallel from tqdm import tqdm from builder import data_builder, model_builder, loss_builder from config.config import load_config_data from dataloader.pc_dataset import get_label_name, update_config from utils.load_save_util import load_checkpoint from utils.metric_util import per_class_iu, fast_hist_crop from utils.trainer_function import Trainer import copy warnings.filterwarnings("ignore") # clear/empty cached memory used by caching allocator torch.cuda.empty_cache() torch.cuda.memory_summary(device=None, abbreviated=False) # training epoch = 0 best_val_miou = 0 global_iter = 0 def main(args): # pytorch_device = torch.device("cuda:2") # torch.device('cuda:2') # os.environ['TORCH_DISTRIBUTED_DEBUG'] = 'true' # os.environ['MASTER_ADDR'] = 'localhost' # os.environ['MASTER_PORT'] = '9994' # os.environ['RANK'] = "0" # If your script expects `--local_rank` argument to be set, please # change it to read from `os.environ['LOCAL_RANK']` instead. # args.local_rank = os.environ['LOCAL_RANK'] os.environ['OMP_NUM_THREADS'] = "1" distributed = False if "WORLD_SIZE" in os.environ: distributed = int(os.environ["WORLD_SIZE"]) > 1 print(f"distributed: {distributed}") pytorch_device = args.local_rank if distributed: torch.cuda.set_device(pytorch_device) torch.distributed.init_process_group(backend='nccl', init_method='env://') args.world_size = torch.distributed.get_world_size() config_path = args.config_path configs = load_config_data(config_path) # send configs parameters to pc_dataset update_config(configs) dataset_config = configs['dataset_params'] train_dataloader_config = configs['train_data_loader'] val_dataloader_config = configs['val_data_loader'] ssl_dataloader_config = configs['ssl_data_loader'] source_val_batch_size = val_dataloader_config['batch_size'] source_train_batch_size = train_dataloader_config['batch_size'] target_train_batch_size = ssl_dataloader_config['batch_size'] model_config = configs['model_params'] train_hypers = configs['train_params'] past_frame = train_hypers['past'] future_frame = train_hypers['future'] ssl = train_hypers['ssl'] grid_size = model_config['output_shape'] num_class = model_config['num_class'] ignore_label = dataset_config['ignore_label'] model_path = train_hypers['model_load_path'] model_path = train_hypers['model_save_path'] SemKITTI_label_name = get_label_name(dataset_config["label_mapping"]) # NB: no ignored class unique_label = np.asarray(sorted(list(SemKITTI_label_name.keys())))[1:] - 1 unique_label_str = [SemKITTI_label_name[x] for x in unique_label + 1] model = model_builder.build(model_config).to(pytorch_device) if os.path.exists(model_path): model = load_checkpoint(model_path, model, map_location=pytorch_device) # if args.mgpus: # student_model = nn.DataParallel(student_model) # #student_model.cuda() # #student_model.cuda() # student_model = student_model().to(pytorch_device) # if args.local_rank >= 1: if distributed: model = DistributedDataParallel( model, device_ids=[pytorch_device], output_device=args.local_rank, find_unused_parameters=False # True ) if ssl: loss_func, lovasz_softmax = loss_builder.build(wce=True, lovasz=True, num_class=num_class, ignore_label=ignore_label, weights=False, ssl=True, fl=False) else: loss_func, lovasz_softmax = loss_builder.build(wce=True, lovasz=True, num_class=num_class, ignore_label=ignore_label, weights=False, fl=False) source_train_dataset_loader, source_val_dataset_loader, _, target_train_dataset_loader = data_builder.build( dataset_config, train_dataloader_config, val_dataloader_config, ssl_dataloader_config=ssl_dataloader_config, grid_size=grid_size, train_hypers=train_hypers) optimizer = optim.Adam(model.parameters(), lr=train_hypers["learning_rate"]) # scheduler = torch.optim.lr_scheduler.OneCycleLR(optimizer_student, max_lr=0.01, # steps_per_epoch=len(source_train_dataset_loader), # epochs=train_hypers["max_num_epochs"]) # global_iter = 0 check_iter = train_hypers['eval_every_n_steps'] global global_iter, best_val_miou, epoch print("|-------------------------Training started-----------------------------------------|") # Define training mode and function trainer = Trainer( model=model, optimizer=optimizer, ckpt_dir=model_path, unique_label=unique_label, unique_label_str=unique_label_str, lovasz_softmax=lovasz_softmax, loss_func=loss_func, ignore_label=ignore_label, train_mode="ema", ssl=ssl, eval_frequency=1, pytorch_device=pytorch_device, warmup_epoch=5, ema_frequency=1) trainer.fit(train_hypers["max_num_epochs"], source_train_dataset_loader, source_train_batch_size, source_val_dataset_loader, source_val_batch_size, test_loader=None, ckpt_save_interval=1, lr_scheduler_each_iter=False) if __name__ == '__main__': # Training settings parser = argparse.ArgumentParser(description='') parser.add_argument('-y', '--config_path', default='config/semantickitti/nuscenes_T3_3.yaml') parser.add_argument('-g', '--mgpus', action='store_true', default=False) parser.add_argument("--local_rank", default=0, type=int) args = parser.parse_args() print(' '.join(sys.argv)) print(args) main(args)
6,366
33.79235
112
py
T-Concord3D
T-Concord3D-master/train.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------| import os import time import argparse import sys import numpy as np import torch import torch.nn as nn import torch.optim as optim from tqdm import tqdm from utils.metric_util import per_class_iu, fast_hist_crop from dataloader.pc_dataset import get_label_name, update_config from builder import data_builder, model_builder, loss_builder from config.config import load_config_data from utils.load_save_util import load_checkpoint import warnings from torch.nn.parallel import DistributedDataParallel warnings.filterwarnings("ignore") # training epoch = 0 best_val_miou = 0 global_iter = 0 def main(args): os.environ['OMP_NUM_THREADS'] = "1" distributed = False if "WORLD_SIZE" in os.environ: distributed = int(os.environ["WORLD_SIZE"]) > 1 print(f"distributed: {distributed}") pytorch_device = args.local_rank if distributed: torch.cuda.set_device(pytorch_device) torch.distributed.init_process_group(backend='nccl', init_method='env://') args.world_size = torch.distributed.get_world_size() config_path = args.config_path configs = load_config_data(config_path) # send config parameters to pc_dataset update_config(configs) dataset_config = configs['dataset_params'] train_dataloader_config = configs['train_data_loader'] val_dataloader_config = configs['val_data_loader'] ssl_dataloader_config = configs['ssl_data_loader'] val_batch_size = val_dataloader_config['batch_size'] train_batch_size = train_dataloader_config['batch_size'] model_config = configs['model_params'] train_hypers = configs['train_params'] past_frame = train_hypers['past'] future_frame = train_hypers['future'] ssl = train_hypers['ssl'] grid_size = model_config['output_shape'] num_class = model_config['num_class'] ignore_label = dataset_config['ignore_label'] model_load_path = train_hypers['model_load_path'] model_save_path = train_hypers['model_save_path'] SemKITTI_label_name = get_label_name(dataset_config["label_mapping"]) unique_label = np.asarray(sorted(list(SemKITTI_label_name.keys())))[1:] - 1 unique_label_str = [SemKITTI_label_name[x] for x in unique_label + 1] my_model = model_builder.build(model_config).to(pytorch_device) if os.path.exists(model_load_path): my_model = load_checkpoint(model_load_path, my_model, map_location=pytorch_device) # if args.mgpus: # my_model = nn.DataParallel(my_model) # #my_model.cuda() # #my_model.cuda() if distributed: my_model = DistributedDataParallel( my_model, device_ids=[pytorch_device], output_device=args.local_rank, find_unused_parameters=True ) # for weighted class loss weighted_class = False # for focal loss focal_loss = False # True # 20 class number of samples from training sample class_weights = np.array([1.40014903e+00, 1.10968683e+00, 5.06321920e+02, 9.19710291e+01, 1.76627589e+01, 1.58902791e+01, 1.49002594e+02, 6.12058299e+02, 1.75137027e+03, 2.47504075e-01, 3.25237847e+00, 3.62211985e-01, 8.77872638e+00, 4.08248861e-01, 9.97997655e-01, 1.91585640e-01, 7.21493239e+00, 4.68076958e-01, 1.69628483e+01, 6.35032127e+01], dtype=np.float32) per_class_weight = None if focal_loss or weighted_class: per_class_weight = torch.from_numpy(class_weights).to(pytorch_device) optimizer = optim.Adam(my_model.parameters(), lr=train_hypers["learning_rate"]) if ssl: loss_func, lovasz_softmax = loss_builder.build(wce=True, lovasz=True, num_class=num_class, ignore_label=ignore_label, weights=per_class_weight, ssl=True, fl=focal_loss) else: loss_func, lovasz_softmax = loss_builder.build(wce=True, lovasz=True, num_class=num_class, ignore_label=ignore_label, weights=per_class_weight, fl=focal_loss) train_dataset_loader, val_dataset_loader, _, _ = data_builder.build(dataset_config, train_dataloader_config, val_dataloader_config, ssl_dataloader_config=ssl_dataloader_config, grid_size=grid_size, train_hypers=train_hypers) class_count = np.zeros(20) my_model.train() # global_iter = 0 check_iter = train_hypers['eval_every_n_steps'] global global_iter, best_val_miou, epoch print("|-------------------------Training started-----------------------------------------|") print(f"focal_loss:{focal_loss}, weighted_cross_entropy: {weighted_class}") while epoch < train_hypers['max_num_epochs']: print(f"epoch: {epoch}") loss_list = [] pbar = tqdm(total=len(train_dataset_loader)) time.sleep(5) # lr_scheduler.step(epoch) def valideting(hist_list, val_loss_list, val_vox_label, val_grid, val_pt_labs, val_pt_fea, ref_st_idx=None, ref_end_idx=None, lcw=None): val_pt_fea_ten = [torch.from_numpy(i).type(torch.FloatTensor).to(pytorch_device) for i in val_pt_fea] val_grid_ten = [torch.from_numpy(i).to(pytorch_device) for i in val_grid] val_label_tensor = val_vox_label.type(torch.LongTensor).to(pytorch_device) predict_labels = my_model(val_pt_fea_ten, val_grid_ten, val_batch_size) # aux_loss = loss_fun(aux_outputs, point_label_tensor) inp = val_label_tensor.size(0) # TODO: check if this is correctly implemented # hack for batch_size mismatch with the number of training example predict_labels = predict_labels[:inp, :, :, :, :] if ssl: lcw_tensor = torch.FloatTensor(lcw).to(pytorch_device) loss = lovasz_softmax(torch.nn.functional.softmax(predict_labels).detach(), val_label_tensor, ignore=ignore_label, lcw=lcw_tensor) + loss_func(predict_labels.detach(), val_label_tensor, lcw=lcw_tensor) else: loss = lovasz_softmax(torch.nn.functional.softmax(predict_labels).detach(), val_label_tensor, ignore=ignore_label) + loss_func(predict_labels.detach(), val_label_tensor) predict_labels = torch.argmax(predict_labels, dim=1) predict_labels = predict_labels.cpu().detach().numpy() for count, i_val_grid in enumerate(val_grid): hist_list.append(fast_hist_crop(predict_labels[ count, val_grid[count][:, 0], val_grid[count][:, 1], val_grid[count][:, 2]], val_pt_labs[count], unique_label)) val_loss_list.append(loss.detach().cpu().numpy()) return hist_list, val_loss_list # if global_iter % check_iter == 0 and epoch >= 1: if epoch >= 1: my_model.eval() hist_list = [] val_loss_list = [] with torch.no_grad(): # Validation with multi-frames and ssl: # if past_frame > 0 and train_hypers['ssl']: for i_iter_val, (_, vox_label, grid, pt_labs, pt_fea, ref_st_idx, ref_end_idx, val_lcw) \ in enumerate(val_dataset_loader): # call the validation and inference with hist_list, val_loss_list = valideting(hist_list, val_loss_list, vox_label, grid, pt_labs, pt_fea, ref_st_idx=ref_st_idx, ref_end_idx=ref_end_idx, lcw=val_lcw) print(f"--------------- epoch: {epoch} ----------------") iou = per_class_iu(sum(hist_list)) print('Validation per class iou: ') for class_name, class_iou in zip(unique_label_str, iou): print('%s : %.2f%%' % (class_name, class_iou * 100)) val_miou = np.nanmean(iou) * 100 # del val_vox_label, val_grid, val_pt_fea # save model if performance is improved if best_val_miou < val_miou: best_val_miou = val_miou if not os.path.exists(model_save_path.split('/')[-2]): os.mkdir(os.path.join(model_save_path.split('/')[-2])) torch.save(my_model.state_dict(), model_save_path) print(f"Current val miou is {np.round(val_miou, 2)} while the best val miou is " f"{np.round(best_val_miou, 2)}") print(f"Current val loss is {np.round(np.mean(val_loss_list), 2)}") def training(i_iter_train, train_vox_label, train_grid, pt_labels, train_pt_fea, ref_st_idx=None, ref_end_idx=None, lcw=None): global global_iter, best_val_miou, epoch train_pt_fea_ten = [torch.from_numpy(i).type(torch.FloatTensor).to(pytorch_device) for i in train_pt_fea] train_vox_ten = [torch.from_numpy(i).to(pytorch_device) for i in train_grid] point_label_tensor = train_vox_label.type(torch.LongTensor).to(pytorch_device) # forward + backward + optimize outputs = my_model(train_pt_fea_ten, train_vox_ten, train_batch_size) inp = point_label_tensor.size(0) # print(f"outputs.size() : {outputs.size()}") # TODO: check if this is correctly implemented # hack for batch_size mismatch with the number of training example outputs = outputs[:inp, :, :, :, :] ################################ if ssl: lcw_tensor = torch.FloatTensor(lcw).to(pytorch_device) loss = lovasz_softmax(torch.nn.functional.softmax(outputs), point_label_tensor, ignore=ignore_label, lcw=lcw_tensor) + loss_func( outputs, point_label_tensor, lcw=lcw_tensor) else: loss = lovasz_softmax(torch.nn.functional.softmax(outputs), point_label_tensor, ignore=ignore_label) + loss_func( outputs, point_label_tensor) # TODO: check --> to mitigate only one element tensors can be converted to Python scalars loss = loss.mean() loss.backward() optimizer.step() loss_list.append(loss.item()) if global_iter % 1000 == 0: if len(loss_list) > 0: print('epoch %d iter %5d, loss: %.3f\n' % (epoch, i_iter_train, np.mean(loss_list))) else: print('loss error') optimizer.zero_grad() global_iter += 1 if global_iter % 100 == 0: pbar.update(100) if global_iter % check_iter == 0: if len(loss_list) > 0: print('epoch %d iter %5d, loss: %.3f\n' % (epoch, i_iter_train, np.mean(loss_list))) else: print('loss error') my_model.train() # training with multi-frames and ssl: # if past_frame > 0 and train_hypers['ssl']: for i_iter_train, (_, vox_label, grid, pt_labs, pt_fea, ref_st_idx, ref_end_idx, lcw) in enumerate( train_dataset_loader): # call the validation and inference with training(i_iter_train, vox_label, grid, pt_labs, pt_fea, ref_st_idx=ref_st_idx, ref_end_idx=ref_end_idx, lcw=lcw) pbar.close() epoch += 1 if __name__ == '__main__': # Training settings parser = argparse.ArgumentParser(description='') parser.add_argument('-y', '--config_path', default='config/semantickitti/nuscenes_S0_0_T11_33_ssl_s20_p80.yaml') parser.add_argument('-g', '--mgpus', action='store_true', default=False) parser.add_argument("--local_rank", default=0, type=int) args = parser.parse_args() print(' '.join(sys.argv)) print(args) main(args)
13,060
42.105611
120
py
T-Concord3D
T-Concord3D-master/tools/rename_kitti_train_pseudo.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------| import os import time import argparse import sys import numpy as np import numpy as np import glob from multiprocessing import Pool import os import shutil import random import math def main(): #sequence = ["04"] #des_seq = ["34"] sequence = ["00", "01", "02","03", "04", "05", "06", "07", "09", "10"] #des_seq = ["11", "12", "13","14", "15", "16", "17", "18", "19", "20"] source = '/mnt/beegfs/gpu/argoverse-tracking-all-training/semantic-kitti/train_pseudo_40/sequences' destination = '/mnt/beegfs/gpu/argoverse-tracking-all-training/semantic-kitti/train_pseudo_40/sequences' for i, sq in enumerate(sequence): files = sorted(glob.glob(os.path.join(source, sq, "velodyne", '*.bin'))) total_frame = range(len(files)) for frame, data in enumerate(files): frame_name = str(frame).zfill(6) frame_data = data[-10:-4] os.rename(os.path.join(source, sq, "velodyne", frame_data + '.bin'), os.path.join( source, sq, "velodyne", frame_name + '.bin')) os.rename(os.path.join(source, sq, "labels", frame_data + '.label'), os.path.join(source, sq, "labels", frame_name + '.label')) # shutil.copy(os.path.join(source, sq, "calib.txt"), os.path.join(destination, des_seq[i], "calib.txt")) # shutil.copy(os.path.join(source, sq, "poses.txt"), os.path.join(destination, des_seq[i], "poses.txt")) # shutil.copy(os.path.join(source, sq, "times.txt"), os.path.join( destination, des_seq[i], "times.txt")) if __name__ == '__main__': main() print(f"------------------------------Task finished-------------------------")
1,739
33.8
140
py
T-Concord3D
T-Concord3D-master/config/config.py
# -*- coding:utf-8 -*- from pathlib import Path from strictyaml import Bool, Float, Int, Map, Seq, Str, as_document, load model_params = Map( { "model_architecture": Str(), "output_shape": Seq(Int()), "fea_dim": Int(), "out_fea_dim": Int(), "num_class": Int(), "num_input_features": Int(), "use_norm": Bool(), "init_size": Int(), } ) dataset_params = Map( { "dataset_type": Str(), "pc_dataset_type": Str(), "ignore_label": Int(), "return_test": Bool(), "fixed_volume_space": Bool(), "label_mapping": Str(), "max_volume_space": Seq(Float()), "min_volume_space": Seq(Float()), } ) train_data_loader = Map( { "data_path": Str(), "imageset": Str(), "return_ref": Bool(), "batch_size": Int(), "shuffle": Bool(), "num_workers": Int(), } ) val_data_loader = Map( { "data_path": Str(), "imageset": Str(), "return_ref": Bool(), "batch_size": Int(), "shuffle": Bool(), "num_workers": Int(), } ) test_data_loader = Map( { "data_path": Str(), "imageset": Str(), "return_ref": Bool(), "batch_size": Int(), "shuffle": Bool(), "num_workers": Int(), } ) ssl_data_loader = Map( { "data_path": Str(), "imageset": Str(), "return_ref": Bool(), "batch_size": Int(), "shuffle": Bool(), "num_workers": Int(), } ) train_params = Map( { "model_load_path": Str(), "model_save_path": Str(), "checkpoint_every_n_steps": Int(), "max_num_epochs": Int(), "eval_every_n_steps": Int(), "learning_rate": Float(), "past": Int(), "future": Int(), "T_past": Str(), "T_future": Str(), "ssl": Bool(), "rgb": Bool(), } ) schema_v4 = Map( { "format_version": Int(), "model_params": model_params, "dataset_params": dataset_params, "train_data_loader": train_data_loader, "val_data_loader": val_data_loader, "test_data_loader": test_data_loader, "ssl_data_loader": ssl_data_loader, "train_params": train_params, } ) SCHEMA_FORMAT_VERSION_TO_SCHEMA = {4: schema_v4} def load_config_data(path: str) -> dict: yaml_string = Path(path).read_text() cfg_without_schema = load(yaml_string, schema=None) schema_version = int(cfg_without_schema["format_version"]) if schema_version not in SCHEMA_FORMAT_VERSION_TO_SCHEMA: raise Exception(f"Unsupported schema format version: {schema_version}.") strict_cfg = load(yaml_string, schema=SCHEMA_FORMAT_VERSION_TO_SCHEMA[schema_version]) cfg: dict = strict_cfg.data return cfg def config_data_to_config(data): # type: ignore return as_document(data, schema_v4) def save_config_data(data: dict, path: str) -> None: cfg_document = config_data_to_config(data) with open(Path(path), "w") as f: f.write(cfg_document.as_yaml())
3,134
22.931298
90
py
T-Concord3D
T-Concord3D-master/config/__init__.py
# -*- coding:utf-8 -*-
23
11
22
py
T-Concord3D
T-Concord3D-master/builder/loss_builder.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------| import torch from utils.lovasz_losses import lovasz_softmax, lovasz_softmax_lcw, cross_entropy_lcw from utils.loss_func import FocalLoss def build(wce=True, lovasz=True, num_class=20, ignore_label=None, weights=None, ssl=False, fl=False): # focal loss and semisupervised learning if ssl and fl: if wce and lovasz: return FocalLoss(weight=weights, ignore_index=ignore_label), lovasz_softmax_lcw elif wce and not lovasz: return wce elif not wce and lovasz: return lovasz_softmax_lcw # only semi-supervised learning if ssl: if wce and lovasz: return cross_entropy_lcw, lovasz_softmax_lcw elif wce and not lovasz: return wce elif not wce and lovasz: return lovasz_softmax_lcw # focal loss on GT (fully supervised) if fl: loss_funs = FocalLoss(weight=weights, ignore_index=ignore_label) else: loss_funs = torch.nn.CrossEntropyLoss(ignore_index=ignore_label) if wce and lovasz: return loss_funs, lovasz_softmax elif wce and not lovasz: return wce elif not wce and lovasz: return lovasz_softmax else: raise NotImplementedError
1,319
30.428571
101
py
T-Concord3D
T-Concord3D-master/builder/model_builder.py
# -*- coding:utf-8 -*- from model.cylinder_3d import get_model_class from model.segment_3d import Asymm_3d_spconv from model.cylinder_feature import cylinder_fea def build(model_config): output_shape = model_config['output_shape'] num_class = model_config['num_class'] num_input_features = model_config['num_input_features'] use_norm = model_config['use_norm'] init_size = model_config['init_size'] fea_dim = model_config['fea_dim'] out_fea_dim = model_config['out_fea_dim'] cylinder_3d_spconv_seg = Asymm_3d_spconv( output_shape=output_shape, use_norm=use_norm, num_input_features=num_input_features, init_size=init_size, nclasses=num_class) cy_fea_net = cylinder_fea(grid_size=output_shape, fea_dim=fea_dim, out_pt_fea_dim=out_fea_dim, fea_compre=num_input_features) model = get_model_class(model_config["model_architecture"])( cylin_model=cy_fea_net, segmentator_spconv=cylinder_3d_spconv_seg, sparse_shape=output_shape ) return model
1,147
30.888889
64
py
T-Concord3D
T-Concord3D-master/builder/data_builder.py
# -*- coding:utf-8 -*- import torch from dataloader.dataset_semantickitti import get_model_class, collate_fn_BEV, collate_fn_BEV_tta from dataloader.pc_dataset import get_pc_model_class def build(dataset_config, train_dataloader_config, val_dataloader_config, test_dataloader_config=None, ssl_dataloader_config=None, grid_size=[480, 360, 32], use_tta=False, train_hypers=None): train_data_path = train_dataloader_config["data_path"] train_imageset = train_dataloader_config["imageset"] val_data_path = val_dataloader_config["data_path"] val_imageset = val_dataloader_config["imageset"] train_ref = train_dataloader_config["return_ref"] val_ref = val_dataloader_config["return_ref"] if test_dataloader_config is not None: test_data_path = test_dataloader_config["data_path"] test_imageset = test_dataloader_config["imageset"] test_ref = test_dataloader_config["return_ref"] # ssl data path for Semi-Supervised training ssl_data_path = None if ssl_dataloader_config is not None: ssl_data_path = ssl_dataloader_config["data_path"] ssl_imageset = ssl_dataloader_config["imageset"] ssl_ref = ssl_dataloader_config["return_ref"] label_mapping = dataset_config["label_mapping"] SemKITTI = get_pc_model_class(dataset_config['pc_dataset_type']) nusc = None if "nusc" in dataset_config['pc_dataset_type']: from nuscenes import NuScenes nusc = NuScenes(version='v1.0-trainval', dataroot=train_data_path, verbose=True) # if we want to train in SSL mode if train_hypers and ssl_dataloader_config and train_hypers['ssl']: train_pt_dataset = SemKITTI(train_data_path, imageset=train_imageset, return_ref=train_ref, label_mapping=label_mapping, train_hypers=train_hypers, ssl_data_path=ssl_data_path) else: train_pt_dataset = SemKITTI(train_data_path, imageset=train_imageset, return_ref=train_ref, label_mapping=label_mapping, train_hypers=train_hypers, ssl_data_path=None) val_pt_dataset = SemKITTI(val_data_path, imageset=val_imageset, return_ref=val_ref, label_mapping=label_mapping, train_hypers=train_hypers) if test_dataloader_config is not None: test_pt_dataset = SemKITTI(test_data_path, imageset=test_imageset, return_ref=test_ref, label_mapping=label_mapping, train_hypers=train_hypers) if ssl_dataloader_config is not None: ssl_pt_dataset = SemKITTI(ssl_data_path, imageset=ssl_imageset, return_ref=ssl_ref, label_mapping=label_mapping, train_hypers=train_hypers) train_dataset = get_model_class(dataset_config['dataset_type'])( train_pt_dataset, grid_size=grid_size, flip_aug=True, fixed_volume_space=dataset_config['fixed_volume_space'], max_volume_space=dataset_config['max_volume_space'], min_volume_space=dataset_config['min_volume_space'], ignore_label=dataset_config["ignore_label"], rotate_aug=True, scale_aug=True, transform_aug=True ) if use_tta: val_dataset = get_model_class(dataset_config['dataset_type'])( val_pt_dataset, grid_size=grid_size, fixed_volume_space=dataset_config['fixed_volume_space'], max_volume_space=dataset_config['max_volume_space'], min_volume_space=dataset_config['min_volume_space'], ignore_label=dataset_config["ignore_label"], rotate_aug=True, scale_aug=True, return_test=True, use_tta=True ) collate_fn_BEV_tmp = collate_fn_BEV_tta else: val_dataset = get_model_class(dataset_config['dataset_type'])( val_pt_dataset, grid_size=grid_size, fixed_volume_space=dataset_config['fixed_volume_space'], max_volume_space=dataset_config['max_volume_space'], min_volume_space=dataset_config['min_volume_space'], ignore_label=dataset_config["ignore_label"], ) collate_fn_BEV_tmp = collate_fn_BEV if use_tta: if test_dataloader_config is not None: test_dataset = get_model_class(dataset_config['dataset_type'])( test_pt_dataset, grid_size=grid_size, fixed_volume_space=dataset_config['fixed_volume_space'], max_volume_space=dataset_config['max_volume_space'], min_volume_space=dataset_config['min_volume_space'], ignore_label=dataset_config["ignore_label"], rotate_aug=True, scale_aug=True, return_test=True, use_tta=True ) collate_fn_BEV_tmp = collate_fn_BEV_tta else: if test_dataloader_config is not None: test_dataset = get_model_class(dataset_config['dataset_type'])( test_pt_dataset, grid_size=grid_size, fixed_volume_space=dataset_config['fixed_volume_space'], max_volume_space=dataset_config['max_volume_space'], min_volume_space=dataset_config['min_volume_space'], ignore_label=dataset_config["ignore_label"], ) collate_fn_BEV_tmp = collate_fn_BEV if ssl_dataloader_config is not None: ssl_dataset = get_model_class(dataset_config['dataset_type'])( ssl_pt_dataset, grid_size=grid_size, fixed_volume_space=dataset_config['fixed_volume_space'], max_volume_space=dataset_config['max_volume_space'], min_volume_space=dataset_config['min_volume_space'], ignore_label=dataset_config["ignore_label"], ) train_dataset_loader = torch.utils.data.DataLoader(dataset=train_dataset, batch_size=train_dataloader_config["batch_size"], collate_fn=collate_fn_BEV, shuffle=train_dataloader_config["shuffle"], num_workers=train_dataloader_config["num_workers"]) val_dataset_loader = torch.utils.data.DataLoader(dataset=val_dataset, batch_size=val_dataloader_config["batch_size"], collate_fn=collate_fn_BEV_tmp, shuffle=val_dataloader_config["shuffle"], num_workers=val_dataloader_config["num_workers"]) test_dataset_loader = None if test_dataloader_config is not None: test_dataset_loader = torch.utils.data.DataLoader(dataset=test_dataset, batch_size=test_dataloader_config["batch_size"], collate_fn=collate_fn_BEV_tmp, shuffle=test_dataloader_config["shuffle"], num_workers=test_dataloader_config["num_workers"]) ssl_dataset_loader = None if ssl_dataloader_config is not None: ssl_dataset_loader = torch.utils.data.DataLoader(dataset=ssl_dataset, batch_size=ssl_dataloader_config["batch_size"], collate_fn=collate_fn_BEV, shuffle=ssl_dataloader_config["shuffle"], num_workers=ssl_dataloader_config["num_workers"]) return train_dataset_loader, val_dataset_loader, test_dataset_loader, ssl_dataset_loader
8,112
48.469512
111
py
T-Concord3D
T-Concord3D-master/builder/__init__.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------|
82
19.75
29
py
T-Concord3D
T-Concord3D-master/utils/trainer_function.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # at 8/10/22 # --------------------------| import argparse import os import sys import time import warnings import numpy as np import torch import torch.optim as optim from torch.nn.parallel import DistributedDataParallel from tqdm import tqdm from builder import data_builder, model_builder, loss_builder from config.config import load_config_data from dataloader.pc_dataset import get_label_name, update_config from utils.load_save_util import load_checkpoint from utils.metric_util import per_class_iu, fast_hist_crop import copy def yield_target_dataset_loader(n_epochs, target_train_dataset_loader): for e in range(n_epochs): for i_iter_train, (_, train_vox_label, train_grid, _, train_pt_fea, ref_st_idx, ref_end_idx, lcw) \ in enumerate(target_train_dataset_loader): yield train_vox_label, train_grid, train_pt_fea, ref_st_idx, ref_end_idx, lcw class Trainer(object): def __init__(self, model, optimizer, ckpt_dir, unique_label, unique_label_str, lovasz_softmax, loss_func, ignore_label, train_mode=None, ssl=None, eval_frequency=1, pytorch_device=0, warmup_epoch=1, ema_frequency=5): self.model = model self.optimizer = optimizer self.model_save_path = ckpt_dir self.unique_label = unique_label self.unique_label_str = unique_label_str self.eval_frequency = eval_frequency self.lovasz_softmax = lovasz_softmax self.loss_func = loss_func self.ignore_label = ignore_label self.train_mode = train_mode self.ssl = ssl self.pytorch_device = pytorch_device self.warmup_epoch = warmup_epoch self.ema_frequency = ema_frequency self.val = False self.best_val_miou = 0 self.progress_value = 100 def criterion(self, outputs, point_label_tensor, lcw=None): if self.ssl: lcw_tensor = torch.FloatTensor(lcw).to(self.pytorch_device) loss = self.lovasz_softmax(torch.nn.functional.softmax(outputs), point_label_tensor, ignore=self.ignore_label, lcw=lcw_tensor) \ + self.loss_func(outputs, point_label_tensor, lcw=lcw_tensor) else: loss = self.lovasz_softmax(torch.nn.functional.softmax(outputs), point_label_tensor, ignore=self.ignore_label) \ + self.loss_func(outputs, point_label_tensor) return loss def validate(self, my_model, val_dataset_loader, val_batch_size, test_loader=None, ssl=None): hist_list = [] val_loss_list = [] my_model.eval() with torch.no_grad(): for i_iter_val, ( _, val_vox_label, val_grid, val_pt_labs, val_pt_fea, ref_st_idx, ref_end_idx, lcw) in enumerate( val_dataset_loader): val_pt_fea_ten = [torch.from_numpy(i).type(torch.FloatTensor).to(self.pytorch_device) for i in val_pt_fea] val_grid_ten = [torch.from_numpy(i).to(self.pytorch_device) for i in val_grid] val_label_tensor = val_vox_label.type(torch.LongTensor).to(self.pytorch_device) predict_labels = my_model(val_pt_fea_ten, val_grid_ten, val_batch_size) # aux_loss = loss_fun(aux_outputs, point_label_tensor) inp = val_label_tensor.size(0) # TODO: check if this is correctly implemented # hack for batch_size mismatch with the number of training example predict_labels = predict_labels[:inp, :, :, :, :] # loss = self.criterion(predict_labels, val_label_tensor, lcw) predict_labels = torch.argmax(predict_labels, dim=1) predict_labels = predict_labels.cpu().detach().numpy() for count, i_val_grid in enumerate(val_grid): hist_list.append(fast_hist_crop(predict_labels[ count, val_grid[count][:, 0], val_grid[count][:, 1], val_grid[count][:, 2]], val_pt_labs[count], self.unique_label)) # val_loss_list.append(loss.detach().cpu().numpy()) return hist_list, val_loss_list def fit(self, n_epochs, source_train_dataset_loader, train_batch_size, val_dataset_loader, val_batch_size, test_loader=None, ckpt_save_interval=1, lr_scheduler_each_iter=False): global_iter = 1 best_val_miou = 0 for epoch in range(n_epochs): pbar = tqdm(total=len(source_train_dataset_loader)) # train the model loss_list = [] self.model.train() # training with multi-frames and ssl: for i_iter_train, ( _, train_vox_label, train_grid, _, train_pt_fea, ref_st_idx, ref_end_idx, lcw) in enumerate( source_train_dataset_loader): # call the validation and inference with train_pt_fea_ten = [torch.from_numpy(i).type(torch.FloatTensor).to(self.pytorch_device) for i in train_pt_fea] # train_grid_ten = [torch.from_numpy(i[:,:2]).to(self.pytorch_device) for i in train_grid] train_vox_ten = [torch.from_numpy(i).to(self.pytorch_device) for i in train_grid] point_label_tensor = train_vox_label.type(torch.LongTensor).to(self.pytorch_device) # forward + backward + optimize outputs = self.model(train_pt_fea_ten, train_vox_ten, train_batch_size) inp = point_label_tensor.size(0) # print(f"outputs.size() : {outputs.size()}") # TODO: check if this is correctly implemented # hack for batch_size mismatch with the number of training example outputs = outputs[:inp, :, :, :, :] ################################ loss = self.criterion(outputs, point_label_tensor, lcw) # TODO: check --> to mitigate only one element tensors can be converted to Python scalars loss = loss.mean() loss.backward() self.optimizer.step() self.optimizer.zero_grad() # Uncomment to use the learning rate scheduler # scheduler.step() loss_list.append(loss.item()) if global_iter % self.progress_value == 0: pbar.update(self.progress_value) if len(loss_list) > 0: print('epoch %d iter %5d, loss: %.3f\n' % (epoch, i_iter_train, np.mean(loss_list))) else: print('loss error') global_iter += 1 # ----------------------------------------------------------------------# # Evaluation/validation with torch.no_grad(): hist_list, val_loss_list = self.validate(self.model, val_dataset_loader, val_batch_size, test_loader, self.ssl) # ----------------------------------------------------------------------# # Print validation mIoU and Loss print(f"--------------- epoch: {epoch} ----------------") iou = per_class_iu(sum(hist_list)) print('Validation per class iou: ') for class_name, class_iou in zip(self.unique_label_str, iou): print('%s : %.2f%%' % (class_name, class_iou * 100)) val_miou = np.nanmean(iou) * 100 # del val_vox_label, val_grid, val_pt_fea # save model if performance is improved if best_val_miou < val_miou: best_val_miou = val_miou torch.save(self.model.state_dict(), self.model_save_path) print('Current val miou is %.3f while the best val miou is %.3f' % (val_miou, best_val_miou)) # print('Current val loss is %.3f' % (np.mean(val_loss_list)))
8,482
42.953368
116
py
T-Concord3D
T-Concord3D-master/utils/load_save_util.py
# -*- coding:utf-8 -*- import torch def load_checkpoint(model_load_path, model, map_location=None): my_model_dict = model.state_dict() if map_location is not None: pre_weight = torch.load(model_load_path, map_location=f'cuda:{map_location}') else: pre_weight = torch.load(model_load_path) part_load = {} match_size = 0 nomatch_size = 0 for k in pre_weight.keys(): value = pre_weight[k] if k[:7] == 'module.': k=k[7:] if k in my_model_dict and my_model_dict[k].shape == value.shape: #print("loading ", k) match_size += 1 part_load[k] = value else: nomatch_size += 1 print("matched parameter sets: {}, and no matched: {}".format(match_size, nomatch_size)) my_model_dict.update(part_load) model.load_state_dict(my_model_dict) return model def load_checkpoint_1b1(model_load_path, model): my_model_dict = model.state_dict() pre_weight = torch.load(model_load_path) part_load = {} match_size = 0 nomatch_size = 0 pre_weight_list = [*pre_weight] my_model_dict_list = [*my_model_dict] for idx in range(len(pre_weight_list)): key_ = pre_weight_list[idx] key_2 = my_model_dict_list[idx] value_ = pre_weight[key_] if my_model_dict[key_2].shape == pre_weight[key_].shape: # print("loading ", k) match_size += 1 part_load[key_2] = value_ else: print(key_) print(key_2) nomatch_size += 1 print("matched parameter sets: {}, and no matched: {}".format(match_size, nomatch_size)) my_model_dict.update(part_load) model.load_state_dict(my_model_dict) return model
1,772
26.703125
92
py
T-Concord3D
T-Concord3D-master/utils/metric_util.py
# -*- coding:utf-8 -*- import numpy as np def fast_hist(pred, label, n): k = (label >= 0) & (label < n) bin_count = np.bincount( n * label[k].astype(int) + pred[k], minlength=n ** 2) return bin_count[:n ** 2].reshape(n, n) def per_class_iu(hist): return np.diag(hist) / (hist.sum(1) + hist.sum(0) - np.diag(hist)) def fast_hist_crop(output, target, unique_label): hist = fast_hist(output.flatten(), target.flatten(), np.max(unique_label) + 2) hist = hist[unique_label + 1, :] hist = hist[:, unique_label + 1] return hist # TODO: check if this implemented correctly def fast_ups_crop(uncrt, target, unique_label): hist = [np.sum(uncrt[target==i]) for i in range(20)] va, cla_count = np.unique(target, return_counts=True) class_count = np.zeros(20) class_count[va] = cla_count return hist, class_count
868
27.966667
82
py
T-Concord3D
T-Concord3D-master/utils/__init__.py
# -*- coding:utf-8 -*- # author: Xinge # @file: __init__.py.py
64
15.25
24
py
T-Concord3D
T-Concord3D-master/utils/ups.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------| def enable_dropout(model): for m in model.modules(): if m.__class__.__name__.startswith('Dropout'): m.train()
218
20.9
54
py
T-Concord3D
T-Concord3D-master/utils/lovasz_losses.py
# -*- coding:utf-8 -*- # author: Xinge """ Lovasz-Softmax and Jaccard hinge loss in PyTorch Maxim Berman 2018 ESAT-PSI KU Leuven (MIT License) """ from __future__ import print_function, division import torch from torch.autograd import Variable import torch.nn.functional as F import numpy as np try: from itertools import ifilterfalse except ImportError: # py3k from itertools import filterfalse as ifilterfalse def lovasz_grad(gt_sorted): """ Computes gradient of the Lovasz extension w.r.t sorted errors See Alg. 1 in paper """ p = len(gt_sorted) gts = gt_sorted.sum() intersection = gts - gt_sorted.float().cumsum(0) union = gts + (1 - gt_sorted).float().cumsum(0) jaccard = 1. - intersection / union if p > 1: # cover 1-pixel case jaccard[1:p] = jaccard[1:p] - jaccard[0:-1] return jaccard def iou_binary(preds, labels, EMPTY=1., ignore=None, per_image=True): """ IoU for foreground class binary: 1 foreground, 0 background """ if not per_image: preds, labels = (preds,), (labels,) ious = [] for pred, label in zip(preds, labels): intersection = ((label == 1) & (pred == 1)).sum() union = ((label == 1) | ((pred == 1) & (label != ignore))).sum() if not union: iou = EMPTY else: iou = float(intersection) / float(union) ious.append(iou) iou = mean(ious) # mean accross images if per_image return 100 * iou def iou(preds, labels, C, EMPTY=1., ignore=None, per_image=False): """ Array of IoU for each (non ignored) class """ if not per_image: preds, labels = (preds,), (labels,) ious = [] for pred, label in zip(preds, labels): iou = [] for i in range(C): if i != ignore: # The ignored label is sometimes among predicted classes (ENet - CityScapes) intersection = ((label == i) & (pred == i)).sum() union = ((label == i) | ((pred == i) & (label != ignore))).sum() if not union: iou.append(EMPTY) else: iou.append(float(intersection) / float(union)) ious.append(iou) ious = [mean(iou) for iou in zip(*ious)] # mean accross images if per_image return 100 * np.array(ious) # --------------------------- BINARY LOSSES --------------------------- def lovasz_hinge(logits, labels, per_image=True, ignore=None): """ Binary Lovasz hinge loss logits: [B, H, W] Variable, logits at each pixel (between -\infty and +\infty) labels: [B, H, W] Tensor, binary ground truth masks (0 or 1) per_image: compute the loss per image instead of per batch ignore: void class id """ if per_image: loss = mean(lovasz_hinge_flat(*flatten_binary_scores(log.unsqueeze(0), lab.unsqueeze(0), ignore)) for log, lab in zip(logits, labels)) else: loss = lovasz_hinge_flat(*flatten_binary_scores(logits, labels, ignore)) return loss def lovasz_hinge_flat(logits, labels): """ Binary Lovasz hinge loss logits: [P] Variable, logits at each prediction (between -\infty and +\infty) labels: [P] Tensor, binary ground truth labels (0 or 1) ignore: label to ignore """ if len(labels) == 0: # only void pixels, the gradients should be 0 return logits.sum() * 0. signs = 2. * labels.float() - 1. errors = (1. - logits * Variable(signs)) errors_sorted, perm = torch.sort(errors, dim=0, descending=True) perm = perm.data gt_sorted = labels[perm] grad = lovasz_grad(gt_sorted) loss = torch.dot(F.relu(errors_sorted), Variable(grad)) return loss def flatten_binary_scores(scores, labels, ignore=None): """ Flattens predictions in the batch (binary case) Remove labels equal to 'ignore' """ scores = scores.view(-1) labels = labels.view(-1) if ignore is None: return scores, labels valid = (labels != ignore) vscores = scores[valid] vlabels = labels[valid] return vscores, vlabels class StableBCELoss(torch.nn.modules.Module): def __init__(self): super(StableBCELoss, self).__init__() def forward(self, input, target): neg_abs = - input.abs() loss = input.clamp(min=0) - input * target + (1 + neg_abs.exp()).log() return loss.mean() def binary_xloss(logits, labels, ignore=None): """ Binary Cross entropy loss logits: [B, H, W] Variable, logits at each pixel (between -\infty and +\infty) labels: [B, H, W] Tensor, binary ground truth masks (0 or 1) ignore: void class id """ logits, labels = flatten_binary_scores(logits, labels, ignore) loss = StableBCELoss()(logits, Variable(labels.float())) return loss # --------------------------- MULTICLASS LOSSES --------------------------- def lovasz_softmax(probas, labels, classes='present', per_image=False, ignore=None): """ Multi-class Lovasz-Softmax loss probas: [B, C, H, W] Variable, class probabilities at each prediction (between 0 and 1). Interpreted as binary (sigmoid) output with outputs of size [B, H, W]. labels: [B, H, W] Tensor, ground truth labels (between 0 and C - 1) classes: 'all' for all, 'present' for classes present in labels, or a list of classes to average. per_image: compute the loss per image instead of per batch ignore: void class labels """ if per_image: loss = mean(lovasz_softmax_flat(*flatten_probas(prob.unsqueeze(0), lab.unsqueeze(0), ignore), classes=classes) for prob, lab in zip(probas, labels)) else: loss = lovasz_softmax_flat(*flatten_probas(probas, labels, ignore), classes=classes) return loss def lovasz_softmax_flat(probas, labels, classes='present'): """ Multi-class Lovasz-Softmax loss probas: [P, C] Variable, class probabilities at each prediction (between 0 and 1) labels: [P] Tensor, ground truth labels (between 0 and C - 1) classes: 'all' for all, 'present' for classes present in labels, or a list of classes to average. """ if probas.numel() == 0: # only void pixels, the gradients should be 0 return probas * 0. C = probas.size(1) losses = [] class_to_sum = list(range(C)) if classes in ['all', 'present'] else classes for c in class_to_sum: fg = (labels == c).float() # foreground for class c if (classes is 'present' and fg.sum() == 0): continue if C == 1: if len(classes) > 1: raise ValueError('Sigmoid output possible only with 1 class') class_pred = probas[:, 0] else: class_pred = probas[:, c] errors = (Variable(fg) - class_pred).abs() errors_sorted, perm = torch.sort(errors, 0, descending=True) perm = perm.data fg_sorted = fg[perm] losses.append(torch.dot(errors_sorted, Variable(lovasz_grad(fg_sorted)))) return mean(losses) def flatten_probas(probas, labels, ignore=None): """ Flattens predictions in the batch """ if probas.dim() == 3: # assumes output of a sigmoid layer B, H, W = probas.size() probas = probas.view(B, 1, H, W) elif probas.dim() == 5: #3D segmentation B, C, L, H, W = probas.size() probas = probas.contiguous().view(B, C, L, H*W) B, C, H, W = probas.size() probas = probas.permute(0, 2, 3, 1).contiguous().view(-1, C) # B * H * W, C = P, C labels = labels.view(-1) if ignore is None: return probas, labels valid = (labels != ignore) vprobas = probas[valid.nonzero().squeeze()] vlabels = labels[valid] return vprobas, vlabels #--------------------------------------------------------------------------------------------------------------------# #---------------------- segmentation confidence probability waited loss function---------------------# def cross_entropy_lcw(probas, labels, ignore_label=0, weights=None, lcw=None): raw_loss_funs = torch.nn.CrossEntropyLoss(ignore_index=ignore_label, reduction='none') # weight label as GT and pseudo #los_func = torch.nn.CrossEntropyLoss(ignore_index=ignore_label) if lcw is not None: #loss = los_func(probas, labels) norm_lcw = (lcw/100.0) raw_loss = raw_loss_funs(probas, labels) weighted_loss = (raw_loss * lcw).mean() else: raise "error: per label weight is none" return weighted_loss def lovasz_softmax_lcw(probas, labels, classes='present', per_image=False, ignore=None, lcw=None): """ Multi-class Lovasz-Softmax loss probas: [B, C, H, W] Variable, class probabilities at each prediction (between 0 and 1). Interpreted as binary (sigmoid) output with outputs of size [B, H, W]. labels: [B, H, W] Tensor, ground truth labels (between 0 and C - 1) classes: 'all' for all, 'present' for classes present in labels, or a list of classes to average. per_image: compute the loss per image instead of per batch ignore: void class labels """ if per_image: loss = mean(lovasz_softmax_flat_lcw(*flatten_probas_lcw(prob.unsqueeze(0), lab.unsqueeze(0), ignore, lcw), classes=classes) for prob, lab in zip(probas, labels)) else: loss = lovasz_softmax_flat_lcw(*flatten_probas_lcw(probas, labels, ignore, lcw), classes=classes) return loss def lovasz_softmax_flat_lcw(probas, labels, lcw=None, classes='present'): """ Multi-class Lovasz-Softmax loss probas: [P, C] Variable, class probabilities at each prediction (between 0 and 1) labels: [P] Tensor, ground truth labels (between 0 and C - 1) classes: 'all' for all, 'present' for classes present in labels, or a list of classes to average. """ if probas.numel() == 0: # only void pixels, the gradients should be 0 return probas * 0. C = probas.size(1) losses = [] class_to_sum = list(range(C)) if classes in ['all', 'present'] else classes for c in class_to_sum: fg = (labels == c).float() # foreground for class c if (classes is 'present' and fg.sum() == 0): continue if C == 1: if len(classes) > 1: raise ValueError('Sigmoid output possible only with 1 class') class_pred = probas[:, 0] else: class_pred = probas[:, c] errors = (Variable(fg) - class_pred).abs() # multiply loss/error by the confidence probability norm_lcw = (lcw/100.0) errors *= norm_lcw errors_sorted, perm = torch.sort(errors, 0, descending=True) perm = perm.data fg_sorted = fg[perm] losses.append(torch.dot(errors_sorted, Variable(lovasz_grad(fg_sorted)))) return mean(losses) def flatten_probas_lcw(probas, labels, ignore=None, lcw=None): """ Flattens predictions in the batch """ if probas.dim() == 3: # assumes output of a sigmoid layer B, H, W = probas.size() probas = probas.view(B, 1, H, W) elif probas.dim() == 5: #3D segmentation B, C, L, H, W = probas.size() probas = probas.contiguous().view(B, C, L, H*W) B, C, H, W = probas.size() probas = probas.permute(0, 2, 3, 1).contiguous().view(-1, C) # B * H * W, C = P, C labels = labels.view(-1) lcw = lcw.view(-1) if ignore is None: return probas, labels, lcw valid = (labels != ignore) vprobas = probas[valid.nonzero().squeeze()] vlabels = labels[valid] vlcw = lcw[valid] return vprobas, vlabels, vlcw #---------------------End of prediction confidence weighted lovasz_softmax loss ----------------------------# #--------------------------------------------------------------------------------------------------------------------# def xloss(logits, labels, ignore=None): """ Cross entropy loss """ return F.cross_entropy(logits, Variable(labels), ignore_index=255) def jaccard_loss(probas, labels,ignore=None, smooth = 100, bk_class = None): """ Something wrong with this loss Multi-class Lovasz-Softmax loss probas: [B, C, H, W] Variable, class probabilities at each prediction (between 0 and 1). Interpreted as binary (sigmoid) output with outputs of size [B, H, W]. labels: [B, H, W] Tensor, ground truth labels (between 0 and C - 1) classes: 'all' for all, 'present' for classes present in labels, or a list of classes to average. per_image: compute the loss per image instead of per batch ignore: void class labels """ vprobas, vlabels = flatten_probas(probas, labels, ignore) true_1_hot = torch.eye(vprobas.shape[1])[vlabels] if bk_class: one_hot_assignment = torch.ones_like(vlabels) one_hot_assignment[vlabels == bk_class] = 0 one_hot_assignment = one_hot_assignment.float().unsqueeze(1) true_1_hot = true_1_hot*one_hot_assignment true_1_hot = true_1_hot.to(vprobas.device) intersection = torch.sum(vprobas * true_1_hot) cardinality = torch.sum(vprobas + true_1_hot) loss = (intersection + smooth / (cardinality - intersection + smooth)).mean() return (1-loss)*smooth def hinge_jaccard_loss(probas, labels,ignore=None, classes = 'present', hinge = 0.1, smooth =100): """ Multi-class Hinge Jaccard loss probas: [B, C, H, W] Variable, class probabilities at each prediction (between 0 and 1). Interpreted as binary (sigmoid) output with outputs of size [B, H, W]. labels: [B, H, W] Tensor, ground truth labels (between 0 and C - 1) classes: 'all' for all, 'present' for classes present in labels, or a list of classes to average. ignore: void class labels """ vprobas, vlabels = flatten_probas(probas, labels, ignore) C = vprobas.size(1) losses = [] class_to_sum = list(range(C)) if classes in ['all', 'present'] else classes for c in class_to_sum: if c in vlabels: c_sample_ind = vlabels == c cprobas = vprobas[c_sample_ind,:] non_c_ind =np.array([a for a in class_to_sum if a != c]) class_pred = cprobas[:,c] max_non_class_pred = torch.max(cprobas[:,non_c_ind],dim = 1)[0] TP = torch.sum(torch.clamp(class_pred - max_non_class_pred, max = hinge)+1.) + smooth FN = torch.sum(torch.clamp(max_non_class_pred - class_pred, min = -hinge)+hinge) if (~c_sample_ind).sum() == 0: FP = 0 else: nonc_probas = vprobas[~c_sample_ind,:] class_pred = nonc_probas[:,c] max_non_class_pred = torch.max(nonc_probas[:,non_c_ind],dim = 1)[0] FP = torch.sum(torch.clamp(class_pred - max_non_class_pred, max = hinge)+1.) losses.append(1 - TP/(TP+FP+FN)) if len(losses) == 0: return 0 return mean(losses) # --------------------------- HELPER FUNCTIONS --------------------------- def isnan(x): return x != x def mean(l, ignore_nan=False, empty=0): """ nanmean compatible with generators. """ l = iter(l) if ignore_nan: l = ifilterfalse(isnan, l) try: n = 1 acc = next(l) except StopIteration: if empty == 'raise': raise ValueError('Empty mean') return empty for n, v in enumerate(l, 2): acc += v if n == 1: return acc return acc / n
15,675
36.864734
131
py
T-Concord3D
T-Concord3D-master/utils/loss_func.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------| import torch import torch.nn as nn import torch.nn.functional as F class FocalLoss(nn.Module): def __init__(self, weight=None, ignore_index=None, gamma=2., reduction='none', ssl=False): nn.Module.__init__(self) self.ignore_index = ignore_index self.weight = weight self.gamma = gamma self.reduction = reduction self.ssl = ssl def forward(self, input_tensor, target_tensor, lcw=None): log_prob = F.log_softmax(input_tensor, dim=1) prob = torch.exp(log_prob) raw_loss = F.nll_loss( ((1 - prob) ** self.gamma) * log_prob, target_tensor, weight=self.weight, reduction=self.reduction, ignore_index=self.ignore_index ) if self.ssl and lcw is not None: norm_lcw = (lcw/100.0) weighted_loss = (raw_loss * lcw).mean() return weighted_loss else: return raw_loss.mean() class WeightedFocalLoss(nn.Module): "Non weighted version of Focal Loss" def __init__(self, weight=None, ignore_index=None, gamma=2., reduction='none', ssl=False): super().__init__() self.ignore_index = ignore_index self.weight = weight self.gamma = gamma self.reduction = reduction self.ssl = ssl def forward(self, inputs, targets): inputs = inputs.squeeze() targets = targets.squeeze() BCE_loss = F.cross_entropy(inputs, targets, reduction='none') pt = torch.exp(-BCE_loss) F_loss = self.weights[targets]*(1-pt)**self.gamma * BCE_loss return F_loss.mean()
1,766
28.949153
69
py
T-Concord3D
T-Concord3D-master/utils/log_util.py
# -*- coding:utf-8 -*- def save_to_log(logdir, logfile, message): f = open(logdir + '/' + logfile, "a") f.write(message + '\n') f.close() return
160
25.833333
42
py
T-Concord3D
T-Concord3D-master/model/cylinder_3d.py
# -*- coding:utf-8 -*- import torch from torch import nn REGISTERED_MODELS_CLASSES = {} def register_model(cls, name=None): global REGISTERED_MODELS_CLASSES if name is None: name = cls.__name__ assert name not in REGISTERED_MODELS_CLASSES, f"exist class: {REGISTERED_MODELS_CLASSES}" REGISTERED_MODELS_CLASSES[name] = cls return cls def get_model_class(name): global REGISTERED_MODELS_CLASSES assert name in REGISTERED_MODELS_CLASSES, f"available class: {REGISTERED_MODELS_CLASSES}" return REGISTERED_MODELS_CLASSES[name] @register_model class cylinder_asym(nn.Module): def __init__(self, cylin_model, segmentator_spconv, sparse_shape, ): super().__init__() self.name = "cylinder_asym" self.cylinder_3d_generator = cylin_model self.cylinder_3d_spconv_seg = segmentator_spconv self.sparse_shape = sparse_shape def forward(self, train_pt_fea_ten, train_vox_ten, batch_size, val_grid=None, voting_num=4, use_tta=False): coords, features_3d = self.cylinder_3d_generator(train_pt_fea_ten, train_vox_ten) # spatial_features = self.cylinder_3d_spconv_seg(features_3d, coords, batch_size) # # return spatial_features if use_tta: batch_size *= voting_num spatial_features = self.cylinder_3d_spconv_seg(features_3d, coords, batch_size) if use_tta: features_ori = torch.split(spatial_features, 1, dim=0) fused_predict = features_ori[0][0, :, val_grid[0][:, 0], val_grid[0][:, 1], val_grid[0][:, 2]] for idx in range(1, voting_num, 1): fused_predict += features_ori[idx][0, :, val_grid[idx][:, 0], val_grid[idx][:, 1], val_grid[idx][:, 2]] return fused_predict else: return spatial_features
1,901
31.793103
119
py
T-Concord3D
T-Concord3D-master/model/segment_3d.py
# -*- coding:utf-8 -*- import numpy as np #import spconv import spconv.pytorch as spconv import torch from torch import nn def conv3x3(in_planes, out_planes, stride=1, indice_key=None): return spconv.SubMConv3d(in_planes, out_planes, kernel_size=3, stride=stride, padding=1, bias=False, indice_key=indice_key) def conv1x3(in_planes, out_planes, stride=1, indice_key=None): return spconv.SubMConv3d(in_planes, out_planes, kernel_size=(1, 3, 3), stride=stride, padding=(0, 1, 1), bias=False, indice_key=indice_key) def conv1x1x3(in_planes, out_planes, stride=1, indice_key=None): return spconv.SubMConv3d(in_planes, out_planes, kernel_size=(1, 1, 3), stride=stride, padding=(0, 0, 1), bias=False, indice_key=indice_key) def conv1x3x1(in_planes, out_planes, stride=1, indice_key=None): return spconv.SubMConv3d(in_planes, out_planes, kernel_size=(1, 3, 1), stride=stride, padding=(0, 1, 0), bias=False, indice_key=indice_key) def conv3x1x1(in_planes, out_planes, stride=1, indice_key=None): return spconv.SubMConv3d(in_planes, out_planes, kernel_size=(3, 1, 1), stride=stride, padding=(1, 0, 0), bias=False, indice_key=indice_key) def conv3x1(in_planes, out_planes, stride=1, indice_key=None): return spconv.SubMConv3d(in_planes, out_planes, kernel_size=(3, 1, 3), stride=stride, padding=(1, 0, 1), bias=False, indice_key=indice_key) def conv1x1(in_planes, out_planes, stride=1, indice_key=None): return spconv.SubMConv3d(in_planes, out_planes, kernel_size=1, stride=stride, padding=1, bias=False, indice_key=indice_key) class ResContextBlock(nn.Module): def __init__(self, in_filters, out_filters, kernel_size=(3, 3, 3), stride=1, indice_key=None): super(ResContextBlock, self).__init__() self.conv1 = conv1x3(in_filters, out_filters, indice_key=indice_key + "bef") self.bn0 = nn.BatchNorm1d(out_filters) self.act1 = nn.LeakyReLU() #elf.conv1_2 = conv3x1(out_filters, out_filters, indice_key=indice_key + "bef") self.conv1_2 = conv1x3(out_filters, out_filters, indice_key=indice_key + "bef") self.bn0_2 = nn.BatchNorm1d(out_filters) self.act1_2 = nn.LeakyReLU() self.conv2 = conv3x1(in_filters, out_filters, indice_key=indice_key + "bef") self.act2 = nn.LeakyReLU() self.bn1 = nn.BatchNorm1d(out_filters) #self.conv3 = conv1x3(out_filters, out_filters, indice_key=indice_key + "bef") self.conv3 = conv3x1(out_filters, out_filters, indice_key=indice_key + "bef") self.act3 = nn.LeakyReLU() self.bn2 = nn.BatchNorm1d(out_filters) self.weight_initialization() def weight_initialization(self): for m in self.modules(): if isinstance(m, nn.BatchNorm1d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) def forward(self, x): shortcut = self.conv1(x) shortcut = shortcut.replace_feature(self.act1(shortcut.features)) shortcut = shortcut.replace_feature(self.bn0(shortcut.features)) shortcut = self.conv1_2(shortcut) shortcut = shortcut.replace_feature(self.act1_2(shortcut.features)) shortcut = shortcut.replace_feature(self.bn0_2(shortcut.features)) resA = self.conv2(x) resA = resA.replace_feature(self.act2(resA.features)) resA = resA.replace_feature(self.bn1(resA.features)) resA = self.conv3(resA) resA = resA.replace_feature(self.act3(resA.features)) resA = resA.replace_feature(self.bn2(resA.features)) resA = resA.replace_feature(resA.features + shortcut.features) return resA class ResBlock(nn.Module): def __init__(self, in_filters, out_filters, dropout_rate, kernel_size=(3, 3, 3), stride=1, pooling=True, drop_out=True, height_pooling=False, indice_key=None): super(ResBlock, self).__init__() self.pooling = pooling #self.drop_out = drop_out self.conv1 = conv3x1(in_filters, out_filters, indice_key=indice_key + "bef") self.act1 = nn.LeakyReLU() self.bn0 = nn.BatchNorm1d(out_filters) # self.conv1_2 = conv1x3(out_filters, out_filters, indice_key=indice_key + "bef") self.conv1_2 = conv3x1(out_filters, out_filters, indice_key=indice_key + "bef") self.act1_2 = nn.LeakyReLU() self.bn0_2 = nn.BatchNorm1d(out_filters) self.conv2 = conv1x3(in_filters, out_filters, indice_key=indice_key + "bef") self.act2 = nn.LeakyReLU() self.bn1 = nn.BatchNorm1d(out_filters) # self.conv3 = conv3x1(out_filters, out_filters, indice_key=indice_key + "bef") self.conv3 = conv1x3(out_filters, out_filters, indice_key=indice_key + "bef") self.act3 = nn.LeakyReLU() self.bn2 = nn.BatchNorm1d(out_filters) if pooling: if height_pooling: self.pool = spconv.SparseConv3d(out_filters, out_filters, kernel_size=3, stride=2, padding=1, indice_key=indice_key, bias=False) else: self.pool = spconv.SparseConv3d(out_filters, out_filters, kernel_size=3, stride=(2, 2, 1), padding=1, indice_key=indice_key, bias=False) self.weight_initialization() def weight_initialization(self): for m in self.modules(): if isinstance(m, nn.BatchNorm1d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) def forward(self, x): shortcut = self.conv1(x) shortcut = shortcut.replace_feature(self.act1(shortcut.features)) shortcut = shortcut.replace_feature(self.bn0(shortcut.features)) shortcut = self.conv1_2(shortcut) shortcut = shortcut.replace_feature(self.act1_2(shortcut.features)) shortcut = shortcut.replace_feature(self.bn0_2(shortcut.features)) resA = self.conv2(x) resA = resA.replace_feature(self.act2(resA.features)) resA = resA.replace_feature(self.bn1(resA.features)) resA = self.conv3(resA) resA = resA.replace_feature(self.act3(resA.features)) resA = resA.replace_feature(self.bn2(resA.features)) resA = resA.replace_feature(resA.features + shortcut.features) if self.pooling: resB = self.pool(resA) return resB, resA else: return resA class UpBlock(nn.Module): def __init__(self, in_filters, out_filters, kernel_size=(3, 3, 3), indice_key=None, up_key=None, dropout_rate=0.25): super(UpBlock, self).__init__() #self.drop_out = drop_out self.trans_dilao = conv3x3(in_filters, out_filters, indice_key=indice_key + "new_up") self.trans_act = nn.LeakyReLU() self.trans_bn = nn.BatchNorm1d(out_filters) self.conv1 = conv1x3(out_filters, out_filters, indice_key=indice_key) self.act1 = nn.LeakyReLU() self.bn1 = nn.BatchNorm1d(out_filters) #self.conv3 = conv3x1(out_filters, out_filters, indice_key=indice_key + "bef") self.conv2 = conv1x3(out_filters, out_filters, indice_key=indice_key) self.act2 = nn.LeakyReLU() self.bn2 = nn.BatchNorm1d(out_filters) #self.conv3 = conv3x3(out_filters, out_filters, indice_key=indice_key) self.conv3 = conv1x3(out_filters, out_filters, indice_key=indice_key) self.act3 = nn.LeakyReLU() self.bn3 = nn.BatchNorm1d(out_filters) # TODO: commnet this drop out after experiment # self.dropout3 = nn.Dropout3d(p=dropout_rate) # added by me self.up_subm = spconv.SparseInverseConv3d(out_filters, out_filters, kernel_size=3, indice_key=up_key, bias=False) self.weight_initialization() def weight_initialization(self): for m in self.modules(): if isinstance(m, nn.BatchNorm1d): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) def forward(self, x, skip): upA = self.trans_dilao(x) upA = upA.replace_feature(self.trans_act(upA.features)) upA = upA.replace_feature(self.trans_bn(upA.features)) ## upsample upA = self.up_subm(upA) upA = upA.replace_feature(upA.features + skip.features) upE = self.conv1(upA) upE = upE.replace_feature(self.act1(upE.features)) upE = upE.replace_feature(self.bn1(upE.features)) upE = self.conv2(upE) upE = upE.replace_feature(self.act2(upE.features)) upE = upE.replace_feature(self.bn2(upE.features)) upE = self.conv3(upE) upE = upE.replace_feature(self.act3(upE.features)) # TODO: comment this drop out after experiment # upE = upE.replace_feature(self.bn3(self.dropout3(upE.features))) # added by me upE = upE.replace_feature(self.bn3(upE.features)) # original implementation return upE class ReconBlock(nn.Module): def __init__(self, in_filters, out_filters, kernel_size=(3, 3, 3), stride=1, indice_key=None): super(ReconBlock, self).__init__() self.conv1 = conv3x1x1(in_filters, out_filters, indice_key=indice_key + "bef") self.bn0 = nn.BatchNorm1d(out_filters) self.act1 = nn.Sigmoid() self.conv1_2 = conv1x3x1(in_filters, out_filters, indice_key=indice_key + "bef") self.bn0_2 = nn.BatchNorm1d(out_filters) self.act1_2 = nn.Sigmoid() self.conv1_3 = conv1x1x3(in_filters, out_filters, indice_key=indice_key + "bef") self.bn0_3 = nn.BatchNorm1d(out_filters) self.act1_3 = nn.Sigmoid() def forward(self, x): shortcut = self.conv1(x) shortcut = shortcut.replace_feature(self.bn0(shortcut.features)) shortcut = shortcut.replace_feature(self.act1(shortcut.features)) shortcut2 = self.conv1_2(x) shortcut2 = shortcut2.replace_feature(self.bn0_2(shortcut2.features)) shortcut2 = shortcut2.replace_feature(self.act1_2(shortcut2.features)) shortcut3 = self.conv1_3(x) shortcut3 = shortcut.replace_feature(self.bn0_3(shortcut3.features)) shortcut3 = shortcut3.replace_feature(self.act1_3(shortcut3.features)) shortcut = shortcut.replace_feature(shortcut.features + shortcut2.features + shortcut3.features) shortcut = shortcut.replace_feature(shortcut.features * x.features) return shortcut class Asymm_3d_spconv(nn.Module): def __init__(self, output_shape, use_norm=True, num_input_features=128, nclasses=20, n_height=32, strict=False, init_size=16): super(Asymm_3d_spconv, self).__init__() self.nclasses = nclasses self.nheight = n_height self.strict = False sparse_shape = np.array(output_shape) # sparse_shape[0] = 11 print(sparse_shape) self.sparse_shape = sparse_shape self.downCntx = ResContextBlock(num_input_features, init_size, indice_key="pre") self.resBlock2 = ResBlock(init_size, 2 * init_size, 0.2, height_pooling=True, indice_key="down2") self.resBlock3 = ResBlock(2 * init_size, 4 * init_size, 0.2, height_pooling=True, indice_key="down3") self.resBlock4 = ResBlock(4 * init_size, 8 * init_size, 0.2, pooling=True, height_pooling=False, indice_key="down4") self.resBlock5 = ResBlock(8 * init_size, 16 * init_size, 0.2, pooling=True, height_pooling=False, indice_key="down5") self.upBlock0 = UpBlock(16 * init_size, 16 * init_size, indice_key="up0", up_key="down5") self.upBlock1 = UpBlock(16 * init_size, 8 * init_size, indice_key="up1", up_key="down4") self.upBlock2 = UpBlock(8 * init_size, 4 * init_size, indice_key="up2", up_key="down3") self.upBlock3 = UpBlock(4 * init_size, 2 * init_size, indice_key="up3", up_key="down2") self.ReconNet = ReconBlock(2 * init_size, 2 * init_size, indice_key="recon") self.logits = spconv.SubMConv3d(4 * init_size, nclasses, indice_key="logit", kernel_size=3, stride=1, padding=1, bias=True) def forward(self, voxel_features, coors, batch_size): coors = coors.int() ret = spconv.SparseConvTensor(voxel_features, coors, self.sparse_shape, batch_size) ret = self.downCntx(ret) down1c, down1b = self.resBlock2(ret) down2c, down2b = self.resBlock3(down1c) down3c, down3b = self.resBlock4(down2c) down4c, down4b = self.resBlock5(down3c) up4e = self.upBlock0(down4c, down4b) up3e = self.upBlock1(up4e, down3b) up2e = self.upBlock2(up3e, down2b) up1e = self.upBlock3(up2e, down1b) up0e = self.ReconNet(up1e) up0e = up0e.replace_feature(torch.cat((up0e.features, up1e.features), 1)) logits = self.logits(up0e) y = logits.dense() return y
13,253
41.07619
121
py
T-Concord3D
T-Concord3D-master/model/__init__.py
# -*- coding:utf-8 -*-
23
11
22
py
T-Concord3D
T-Concord3D-master/model/cylinder_feature.py
# -*- coding:utf-8 -*- import torch import torch.nn as nn import torch.nn.functional as F import numpy as np import numba as nb import multiprocessing import torch_scatter class cylinder_fea(nn.Module): def __init__(self, grid_size, fea_dim=3, out_pt_fea_dim=64, max_pt_per_encode=64, fea_compre=None): super(cylinder_fea, self).__init__() self.PPmodel = nn.Sequential( nn.BatchNorm1d(fea_dim), nn.Linear(fea_dim, 64), nn.BatchNorm1d(64), nn.ReLU(), nn.Linear(64, 128), nn.BatchNorm1d(128), nn.ReLU(), nn.Linear(128, 256), nn.BatchNorm1d(256), nn.ReLU(), nn.Linear(256, out_pt_fea_dim) ) self.max_pt = max_pt_per_encode self.fea_compre = fea_compre self.grid_size = grid_size kernel_size = 3 self.local_pool_op = torch.nn.MaxPool2d(kernel_size, stride=1, padding=(kernel_size - 1) // 2, dilation=1) self.pool_dim = out_pt_fea_dim # point feature compression if self.fea_compre is not None: self.fea_compression = nn.Sequential( nn.Linear(self.pool_dim, self.fea_compre), nn.ReLU()) self.pt_fea_dim = self.fea_compre else: self.pt_fea_dim = self.pool_dim def forward(self, pt_fea, xy_ind): cur_dev = pt_fea[0].get_device() # concate everything cat_pt_ind = [] # for i_batch in range(len(xy_ind)): # cat_pt_ind.append(F.pad(xy_ind[i_batch], (1, 0), 'constant', value=i_batch)) # Awet Optimized append into list comprehension for faster runtime cat_pt_ind = [F.pad(xy_ind[i_batch], (1, 0), 'constant', value=i_batch) for i_batch in range(len(xy_ind)) ] cat_pt_fea = torch.cat(pt_fea, dim=0) cat_pt_ind = torch.cat(cat_pt_ind, dim=0) pt_num = cat_pt_ind.shape[0] # shuffle the data shuffled_ind = torch.randperm(pt_num, device=cur_dev) cat_pt_fea = cat_pt_fea[shuffled_ind, :] cat_pt_ind = cat_pt_ind[shuffled_ind, :] # unique xy grid index unq, unq_inv, unq_cnt = torch.unique(cat_pt_ind, return_inverse=True, return_counts=True, dim=0) unq = unq.type(torch.int64) # process feature processed_cat_pt_fea = self.PPmodel(cat_pt_fea) pooled_data = torch_scatter.scatter_max(processed_cat_pt_fea, unq_inv, dim=0)[0] if self.fea_compre: processed_pooled_data = self.fea_compression(pooled_data) else: processed_pooled_data = pooled_data return unq, processed_pooled_data
2,812
30.965909
115
py
T-Concord3D
T-Concord3D-master/dataloader/dataset_semantickitti.py
# -*- coding:utf-8 -*- """ SemKITTI dataloader """ import os import numpy as np import torch import random import time import numba as nb import yaml from torch.utils import data import pickle REGISTERED_DATASET_CLASSES = {} def register_dataset(cls, name=None): global REGISTERED_DATASET_CLASSES if name is None: name = cls.__name__ assert name not in REGISTERED_DATASET_CLASSES, f"exist class: {REGISTERED_DATASET_CLASSES}" REGISTERED_DATASET_CLASSES[name] = cls return cls def get_model_class(name): global REGISTERED_DATASET_CLASSES assert name in REGISTERED_DATASET_CLASSES, f"available class: {REGISTERED_DATASET_CLASSES}" return REGISTERED_DATASET_CLASSES[name] @register_dataset class voxel_dataset(data.Dataset): def __init__(self, in_dataset, grid_size, rotate_aug=False, flip_aug=False, ignore_label=255, return_test=False, fixed_volume_space=False, max_volume_space=[50, 50, 1.5], min_volume_space=[-50, -50, -3], cut_mix=False): 'Initialization' self.point_cloud_dataset = in_dataset self.grid_size = np.asarray(grid_size) self.rotate_aug = rotate_aug self.ignore_label = ignore_label self.return_test = return_test self.flip_aug = flip_aug self.fixed_volume_space = fixed_volume_space self.max_volume_space = max_volume_space self.min_volume_space = min_volume_space # TODO check if the cut and mix augmentation is implemented correctly self.cut_mix = cut_mix def __len__(self): 'Denotes the total number of samples' return len(self.point_cloud_dataset) def __getitem__(self, index): 'Generates one sample of data' data = self.point_cloud_dataset[index] # initialization xyz = None labels = None sig = None lcw = None ref_st_ind = None ref_end_ind = None if len(data) == 2: xyz, labels = data elif len(data) == 3: xyz, labels, sig = data if len(sig.shape) == 2: sig = np.squeeze(sig) elif len(data) == 5: xyz, labels, sig, ref_st_ind, ref_end_ind = data if len(sig.shape) == 2: sig = np.squeeze(sig) elif len(data) == 6: xyz, labels, sig, lcw, ref_st_ind, ref_end_ind = data else: raise Exception('Return invalid data tuple') # TODO: check -------------------------------- # cut mix data augmentation by grabbing instance and add it to a new scene if self.cut_mix: # load/grab the object dir = '/mnt/beegfs/gpu/argoverse-tracking-all-training/WOD/processed/Labeled/cut_mix' new_xyz = np.load(f"{dir}/pcl.npy") new_label_all = np.load(f"{dir}/ss_id.npy") unique_obj = np.unique(new_label_all[:, 0]) sel_obj_rand = np.random.choice(len(unique_obj), 5) new_label = [] for id in sel_obj_rand: obj_mask = new_label_all[:, 0] == id new_label.append(new_label_all[obj_mask]) new_label = np.concatenate(new_label, axis=0) # perform random flipping and rotation flip_type = np.random.choice(4, 1) if flip_type == 1: new_xyz[:, 0] = -new_xyz[:, 0] elif flip_type == 2: new_xyz[:, 1] = -new_xyz[:, 1] elif flip_type == 3: new_xyz[:, :2] = -new_xyz[:, :2] rotate_rad = np.deg2rad(np.random.random() * 360) c, s = np.cos(rotate_rad), np.sin(rotate_rad) j = np.matrix([[c, s], [-s, c]]) new_xyz[:, :2] = np.dot(new_xyz[:, :2], j) xyz = np.concatenate(xyz, new_xyz[:, :3], axis=0) labels = np.concatenate(labels, new_label[:, 1], axis=0) if sig is not None: sig = np.concatenate(sig, new_xyz[:, 3], axis=0) if lcw is not None: lcw = np.concatenate(lcw, np.ones_like(new_label[:, 1]), axis=0) # random data augmentation by rotation if self.rotate_aug: rotate_rad = np.deg2rad(np.random.random() * 360) c, s = np.cos(rotate_rad), np.sin(rotate_rad) j = np.matrix([[c, s], [-s, c]]) xyz[:, :2] = np.dot(xyz[:, :2], j) # random data augmentation by flip x , y or x+y if self.flip_aug: flip_type = np.random.choice(4, 1) if flip_type == 1: xyz[:, 0] = -xyz[:, 0] elif flip_type == 2: xyz[:, 1] = -xyz[:, 1] elif flip_type == 3: xyz[:, :2] = -xyz[:, :2] max_bound = np.percentile(xyz, 100, axis=0) min_bound = np.percentile(xyz, 0, axis=0) if self.fixed_volume_space: max_bound = np.asarray(self.max_volume_space) min_bound = np.asarray(self.min_volume_space) # get grid index crop_range = max_bound - min_bound cur_grid_size = self.grid_size intervals = crop_range / (cur_grid_size - 1) if (intervals == 0).any(): print("Zero interval!") grid_ind = (np.floor((np.clip(xyz, min_bound, max_bound) - min_bound) / intervals)).astype(np.int) # process voxel position voxel_position = np.zeros(self.grid_size, dtype=np.float32) dim_array = np.ones(len(self.grid_size) + 1, int) dim_array[0] = -1 voxel_position = np.indices(self.grid_size) * intervals.reshape(dim_array) + min_bound.reshape(dim_array) # process labels processed_label = np.ones(self.grid_size, dtype=np.uint8) * self.ignore_label label_voxel_pair = np.concatenate([grid_ind, labels], axis=1) label_voxel_pair = label_voxel_pair[np.lexsort((grid_ind[:, 0], grid_ind[:, 1], grid_ind[:, 2])), :] processed_label = nb_process_label(np.copy(processed_label), label_voxel_pair) # TODO: check if there is lcw label confidence weight if len(data) == 6: # process the lcw processed_lcw = np.ones(self.grid_size, dtype=np.uint8) * self.ignore_label lcw_voxel_pair = np.concatenate([grid_ind, lcw], axis=1) lcw_voxel_pair = lcw_voxel_pair[np.lexsort((grid_ind[:, 0], grid_ind[:, 1], grid_ind[:, 2])), :] processed_lcw = nb_process_label(np.copy(processed_lcw), lcw_voxel_pair) data_tuple = (voxel_position, processed_label) # center data on each voxel for PTnet voxel_centers = (grid_ind.astype(np.float32) + 0.5) * intervals + min_bound return_xyz = xyz - voxel_centers return_xyz = np.concatenate((return_xyz, xyz), axis=1) if len(data) == 2: return_fea = return_xyz elif len(data) >= 3: return_fea = np.concatenate((return_xyz, sig[..., np.newaxis]), axis=1) # np.concatenate((return_xyz, sig), axis=1)# if self.return_test: data_tuple += (grid_ind, labels, return_fea, index) else: data_tuple += (grid_ind, labels, return_fea) if len(data) == 6: data_tuple += (processed_lcw, ref_st_ind, ref_end_ind) elif len(data) == 5: data_tuple += (ref_st_ind, ref_end_ind) return data_tuple # transformation between Cartesian coordinates and polar coordinates def cart2polar(input_xyz): rho = np.sqrt(input_xyz[:, 0] ** 2 + input_xyz[:, 1] ** 2) phi = np.arctan2(input_xyz[:, 1], input_xyz[:, 0]) return np.stack((rho, phi, input_xyz[:, 2]), axis=1) def polar2cat(input_xyz_polar): # print(input_xyz_polar.shape) x = input_xyz_polar[0] * np.cos(input_xyz_polar[1]) y = input_xyz_polar[0] * np.sin(input_xyz_polar[1]) return np.stack((x, y, input_xyz_polar[2]), axis=0) @register_dataset class cylinder_dataset(data.Dataset): def __init__(self, in_dataset, grid_size, rotate_aug=False, flip_aug=False, ignore_label=255, return_test=False, fixed_volume_space=False, max_volume_space=[50, np.pi, 2], min_volume_space=[0, -np.pi, -4], scale_aug=False, transform_aug=False, trans_std=[0.1, 0.1, 0.1], min_rad=-np.pi / 4, max_rad=np.pi / 4, cut_mix=False, use_tta=False): self.point_cloud_dataset = in_dataset self.grid_size = np.asarray(grid_size) self.rotate_aug = rotate_aug self.flip_aug = flip_aug self.scale_aug = scale_aug self.ignore_label = ignore_label self.return_test = return_test self.fixed_volume_space = fixed_volume_space self.max_volume_space = max_volume_space self.min_volume_space = min_volume_space self.transform = transform_aug self.trans_std = trans_std self.cut_mix = cut_mix self.use_tta = use_tta self.noise_rotation = np.random.uniform(min_rad, max_rad) def __len__(self): 'Denotes the total number of samples' return len(self.point_cloud_dataset) def rotation_points_single_angle(self, points, angle, axis=0): # points: [N, 3] rot_sin = np.sin(angle) rot_cos = np.cos(angle) if axis == 1: rot_mat_T = np.array( [[rot_cos, 0, -rot_sin], [0, 1, 0], [rot_sin, 0, rot_cos]], dtype=points.dtype) elif axis == 2 or axis == -1: rot_mat_T = np.array( [[rot_cos, -rot_sin, 0], [rot_sin, rot_cos, 0], [0, 0, 1]], dtype=points.dtype) elif axis == 0: rot_mat_T = np.array( [[1, 0, 0], [0, rot_cos, -rot_sin], [0, rot_sin, rot_cos]], dtype=points.dtype) else: raise ValueError("axis should in range") return points @ rot_mat_T def __getitem__(self, index): 'Generates one sample of data' data = self.point_cloud_dataset[index] if self.use_tta: data_total = [] voting = 4 for idx in range(voting): data_single_ori = self.get_single_sample(data, index, idx) data_total.append(data_single_ori) data_total = tuple(data_total) return data_total else: data_single = self.get_single_sample(data, index) return data_single def get_single_sample(self, data, index, vote_idx=0): split = self.point_cloud_dataset.imageset # initialization xyz = None labels = None sig = None lcw = None ref_st_ind = None ref_end_ind = None if len(data) == 2: xyz, labels = data elif len(data) == 3: xyz, labels, sig = data if len(sig.shape) == 2: sig = np.squeeze(sig) elif len(data) == 5: xyz, labels, sig, ref_st_ind, ref_end_ind = data if len(sig.shape) == 2: sig = np.squeeze(sig) elif len(data) == 6: xyz, labels, sig, lcw, ref_st_ind, ref_end_ind = data if len(sig.shape) == 2: sig = np.squeeze(sig) else: raise Exception('Return invalid data tuple') # TODO: check ----------------------------------------------------- # cut mix data augmentation by grabbing instance and add it to a new scene if self.cut_mix and ((split == 'train') or (split == 'ssl')): # load/grab the object dir = '/mnt/beegfs/gpu/argoverse-tracking-all-training/WOD/processed/Labeled/cut_mix' new_xyz_all = np.load(f"{dir}/pcl.npy") new_label_all = np.load(f"{dir}/ss_id.npy") unique_obj = np.unique(new_label_all[:, 0]) num_object = 10 sel_obj_rand = np.random.choice(len(unique_obj), num_object) aug_label = [] aug_xyz = [] for id in sel_obj_rand: obj_mask = new_label_all[:, 0] == id new_label = new_label_all[obj_mask] new_xyz = new_xyz_all[obj_mask] # perform random mix/placement on the road road_mask = np.squeeze(labels) == 18 road_pcl = xyz[road_mask] mix_pos_rand = np.random.choice(len(road_pcl), 1) mix_position = road_pcl[mix_pos_rand, :] mix_p_x = mix_position[:, 0] - 0.5 mix_p_y = mix_position[:, 1] - 0.5 mix_p_z = mix_position[:, 2] new_xyz[:, 0] = new_xyz[:, 0] - np.max(new_xyz[:, 0]) new_xyz[:, 1] = new_xyz[:, 1] - np.max(new_xyz[:, 1]) new_xyz[:, 2] = new_xyz[:, 2] - np.min(new_xyz[:, 2]) new_xyz[:, 0] = new_xyz[:, 0] + mix_p_x new_xyz[:, 1] = new_xyz[:, 1] + mix_p_y new_xyz[:, 2] = new_xyz[:, 2] + mix_p_z if self.use_tta: flip_type = vote_idx else: flip_type = np.random.choice(4, 1) if flip_type == 1: new_xyz[:, 0] = -new_xyz[:, 0] elif flip_type == 2: new_xyz[:, 1] = -new_xyz[:, 1] rotate_rad = np.deg2rad(np.random.random() * 360) c, s = np.cos(rotate_rad), np.sin(rotate_rad) j = np.matrix([[c, s], [-s, c]]) new_xyz[:, :2] = np.dot(new_xyz[:, :2], j) aug_label.append(new_label) aug_xyz.append(new_xyz) mframe = int(len(new_xyz) / 130000) if mframe > 1: for i in range(1, mframe): new_xyz[:, 0] = new_xyz[:, 0] - i / 2 aug_label.append(new_label) aug_xyz.append(new_xyz) new_label = np.concatenate(aug_label, axis=0) new_xyz = np.concatenate(aug_xyz, axis=0) # combine gt data and cut_mix augmentation xyz = np.concatenate([xyz, new_xyz[:, :3]], axis=0) labels = np.concatenate([labels, new_label[:, 1].reshape(-1, 1)], axis=0) if sig is not None: sig = np.concatenate([sig.reshape(-1, 1), new_xyz[:, 3].reshape(-1, 1)], axis=0) sig = np.squeeze(sig) if lcw is not None: new_lcw = np.ones_like(new_label[:, 1]) * 100 lcw = np.concatenate([lcw, new_lcw.reshape(-1, 1)], axis=0) # random data augmentation by rotation if self.rotate_aug: rotate_rad = np.deg2rad(np.random.random() * 90) - np.pi / 4 c, s = np.cos(rotate_rad), np.sin(rotate_rad) j = np.matrix([[c, s], [-s, c]]) xyz[:, :2] = np.dot(xyz[:, :2], j) # random data augmentation by flip x , y or x+y if self.flip_aug: if self.use_tta: flip_type = vote_idx else: flip_type = np.random.choice(4, 1) if flip_type == 1: xyz[:, 0] = -xyz[:, 0] elif flip_type == 2: xyz[:, 1] = -xyz[:, 1] elif flip_type == 3: xyz[:, :2] = -xyz[:, :2] if self.scale_aug: noise_scale = np.random.uniform(0.95, 1.05) xyz[:, 0] = noise_scale * xyz[:, 0] xyz[:, 1] = noise_scale * xyz[:, 1] # convert coordinate into polar coordinates if self.transform: noise_translate = np.array([np.random.normal(0, self.trans_std[0], 1), np.random.normal(0, self.trans_std[1], 1), np.random.normal(0, self.trans_std[2], 1)]).T xyz[:, 0:3] += noise_translate xyz_pol = cart2polar(xyz) max_bound_r = np.percentile(xyz_pol[:, 0], 100, axis=0) min_bound_r = np.percentile(xyz_pol[:, 0], 0, axis=0) max_bound = np.max(xyz_pol[:, 1:], axis=0) min_bound = np.min(xyz_pol[:, 1:], axis=0) max_bound = np.concatenate(([max_bound_r], max_bound)) min_bound = np.concatenate(([min_bound_r], min_bound)) if self.fixed_volume_space: max_bound = np.asarray(self.max_volume_space) min_bound = np.asarray(self.min_volume_space) # get grid index crop_range = max_bound - min_bound cur_grid_size = self.grid_size intervals = crop_range / (cur_grid_size - 1) if (intervals == 0).any(): print("Zero interval!") grid_ind = (np.floor((np.clip(xyz_pol, min_bound, max_bound) - min_bound) / intervals)).astype(np.int) voxel_position = np.zeros(self.grid_size, dtype=np.float32) dim_array = np.ones(len(self.grid_size) + 1, int) dim_array[0] = -1 voxel_position = np.indices(self.grid_size) * intervals.reshape(dim_array) + min_bound.reshape(dim_array) voxel_position = polar2cat(voxel_position) processed_label = np.ones(self.grid_size, dtype=np.uint8) * self.ignore_label label_voxel_pair = np.concatenate([grid_ind, labels], axis=1) label_voxel_pair = label_voxel_pair[np.lexsort((grid_ind[:, 0], grid_ind[:, 1], grid_ind[:, 2])), :] processed_label = nb_process_label(np.copy(processed_label), label_voxel_pair) # TODO: check if there is lcw label confidence weight if len(data) == 6: processed_lcw = np.ones(self.grid_size, dtype=np.uint8) * self.ignore_label lcw_voxel_pair = np.concatenate([grid_ind, lcw], axis=1) lcw_voxel_pair = lcw_voxel_pair[np.lexsort((grid_ind[:, 0], grid_ind[:, 1], grid_ind[:, 2])), :] processed_lcw = nb_process_label(np.copy(processed_lcw), lcw_voxel_pair) data_tuple = (voxel_position, processed_label) # center data on each voxel for PTnet voxel_centers = (grid_ind.astype(np.float32) + 0.5) * intervals + min_bound return_xyz = xyz_pol - voxel_centers return_xyz = np.concatenate((return_xyz, xyz_pol, xyz[:, :2]), axis=1) if len(data) == 2: return_fea = return_xyz elif len(data) >= 3: return_fea = np.concatenate((return_xyz, sig[..., np.newaxis]), axis=1) # np.concatenate((return_xyz, sig), axis=1) # if self.return_test: data_tuple += (grid_ind, labels, return_fea, index) else: data_tuple += (grid_ind, labels, return_fea) # include reference frame start and end index if len(data) == 6: data_tuple += (processed_lcw, ref_st_ind, ref_end_ind) # include pseudo label confidence weights elif len(data) == 5: data_tuple += (ref_st_ind, ref_end_ind) return data_tuple @register_dataset class polar_dataset(data.Dataset): def __init__(self, in_dataset, grid_size, rotate_aug=False, flip_aug=False, ignore_label=255, return_test=False, fixed_volume_space=False, max_volume_space=[50, np.pi, 2], min_volume_space=[0, -np.pi, -4], scale_aug=False): self.point_cloud_dataset = in_dataset self.grid_size = np.asarray(grid_size) self.rotate_aug = rotate_aug self.flip_aug = flip_aug self.scale_aug = scale_aug self.ignore_label = ignore_label self.return_test = return_test self.fixed_volume_space = fixed_volume_space self.max_volume_space = max_volume_space self.min_volume_space = min_volume_space def __len__(self): 'Denotes the total number of samples' return len(self.point_cloud_dataset) def __getitem__(self, index): 'Generates one sample of data' data = self.point_cloud_dataset[index] if len(data) == 2: xyz, labels = data elif len(data) == 3: xyz, labels, sig = data if len(sig.shape) == 2: sig = np.squeeze(sig) elif len(data) == 5: xyz, labels, sig, ref_st_ind, ref_end_ind = data if len(sig.shape) == 2: sig = np.squeeze(sig) elif len(data) == 6: xyz, labels, sig, lcw, ref_st_ind, ref_end_ind = data if len(sig.shape) == 2: sig = np.squeeze(sig) else: raise Exception('Return invalid data tuple') # random data augmentation by rotation if self.rotate_aug: rotate_rad = np.deg2rad(np.random.random() * 45) - np.pi / 8 c, s = np.cos(rotate_rad), np.sin(rotate_rad) j = np.matrix([[c, s], [-s, c]]) xyz[:, :2] = np.dot(xyz[:, :2], j) # random data augmentation by flip x , y or x+y if self.flip_aug: flip_type = np.random.choice(4, 1) if flip_type == 1: xyz[:, 0] = -xyz[:, 0] elif flip_type == 2: xyz[:, 1] = -xyz[:, 1] elif flip_type == 3: xyz[:, :2] = -xyz[:, :2] if self.scale_aug: noise_scale = np.random.uniform(0.95, 1.05) xyz[:, 0] = noise_scale * xyz[:, 0] xyz[:, 1] = noise_scale * xyz[:, 1] xyz_pol = cart2polar(xyz) max_bound_r = np.percentile(xyz_pol[:, 0], 100, axis=0) min_bound_r = np.percentile(xyz_pol[:, 0], 0, axis=0) max_bound = np.max(xyz_pol[:, 1:], axis=0) min_bound = np.min(xyz_pol[:, 1:], axis=0) max_bound = np.concatenate(([max_bound_r], max_bound)) min_bound = np.concatenate(([min_bound_r], min_bound)) if self.fixed_volume_space: max_bound = np.asarray(self.max_volume_space) min_bound = np.asarray(self.min_volume_space) # get grid index crop_range = max_bound - min_bound cur_grid_size = self.grid_size intervals = crop_range / (cur_grid_size - 1) if (intervals == 0).any(): print("Zero interval!") grid_ind = (np.floor((np.clip(xyz_pol, min_bound, max_bound) - min_bound) / intervals)).astype(np.int) voxel_position = np.zeros(self.grid_size, dtype=np.float32) dim_array = np.ones(len(self.grid_size) + 1, int) dim_array[0] = -1 voxel_position = np.indices(self.grid_size) * intervals.reshape(dim_array) + min_bound.reshape(dim_array) voxel_position = polar2cat(voxel_position) processed_label = np.ones(self.grid_size, dtype=np.uint8) * self.ignore_label label_voxel_pair = np.concatenate([grid_ind, labels], axis=1) label_voxel_pair = label_voxel_pair[np.lexsort((grid_ind[:, 0], grid_ind[:, 1], grid_ind[:, 2])), :] processed_label = nb_process_label(np.copy(processed_label), label_voxel_pair) # TODO: check if there is lcw label confidence weight if len(data) == 6: processed_lcw = np.ones(self.grid_size, dtype=np.uint8) * self.ignore_label lcw_voxel_pair = np.concatenate([grid_ind, lcw], axis=1) lcw_voxel_pair = lcw_voxel_pair[np.lexsort((grid_ind[:, 0], grid_ind[:, 1], grid_ind[:, 2])), :] processed_lcw = nb_process_label(np.copy(processed_lcw), lcw_voxel_pair) data_tuple = (voxel_position, processed_label) # center data on each voxel for PTnet voxel_centers = (grid_ind.astype(np.float32) + 0.5) * intervals + min_bound return_xyz = xyz_pol - voxel_centers return_xyz = np.concatenate((return_xyz, xyz_pol, xyz[:, :2]), axis=1) if len(data) == 2: return_fea = return_xyz elif len(data) >= 3: return_fea = np.concatenate((return_xyz, sig[..., np.newaxis]), axis=1) # np.concatenate((return_xyz, sig), axis=1) # if self.return_test: data_tuple += (grid_ind, labels, return_fea, index) else: data_tuple += (grid_ind, labels, return_fea) # include pseudo label confidence weights if len(data) == 6: data_tuple += (processed_lcw, ref_st_ind, ref_end_ind) # refrence frame index elif len(data) == 5: data_tuple += (ref_st_ind, ref_end_ind) return data_tuple @nb.jit('u1[:,:,:](u1[:,:,:],i8[:,:])', nopython=True, cache=True, parallel=False) def nb_process_label(processed_label, sorted_label_voxel_pair): label_size = 256 counter = np.zeros((label_size,), dtype=np.uint16) counter[sorted_label_voxel_pair[0, 3]] = 1 cur_sear_ind = sorted_label_voxel_pair[0, :3] for i in range(1, sorted_label_voxel_pair.shape[0]): cur_ind = sorted_label_voxel_pair[i, :3] if not np.all(np.equal(cur_ind, cur_sear_ind)): processed_label[cur_sear_ind[0], cur_sear_ind[1], cur_sear_ind[2]] = np.argmax(counter) counter = np.zeros((label_size,), dtype=np.uint16) cur_sear_ind = cur_ind counter[sorted_label_voxel_pair[i, 3]] += 1 processed_label[cur_sear_ind[0], cur_sear_ind[1], cur_sear_ind[2]] = np.argmax(counter) return processed_label @nb.jit('u1[:,:,:](u1[:,:,:],i8[:,:])', nopython=True, cache=True, parallel=False) def nb_process_lcw(processed_label, sorted_label_voxel_pair): label_size = 256 counter = np.zeros((label_size,), dtype=np.float32) counter[sorted_label_voxel_pair[0, 3]] = 1 cur_sear_ind = sorted_label_voxel_pair[0, :3] for i in range(1, sorted_label_voxel_pair.shape[0]): cur_ind = sorted_label_voxel_pair[i, :3] if not np.all(np.equal(cur_ind, cur_sear_ind)): processed_label[cur_sear_ind[0], cur_sear_ind[1], cur_sear_ind[2]] = np.argmax(counter) counter = np.zeros((label_size,), dtype=np.float32) cur_sear_ind = cur_ind counter[sorted_label_voxel_pair[i, 3]] += 1 processed_label[cur_sear_ind[0], cur_sear_ind[1], cur_sear_ind[2]] = np.argmax(counter) return processed_label def collate_fn_BEV(data): data2stack = np.stack([d[0] for d in data]).astype(np.float32) label2stack = np.stack([d[1] for d in data]).astype(np.int) grid_ind_stack = [d[2] for d in data] point_label = [d[3] for d in data] xyz = [d[4] for d in data] ref_st_index = None ref_end_index = None lcw2stack = None # if multi frame but not ssl: also add the start and end index of reference scan/frame if len(data[0]) == 7: ref_st_index = [d[5] for d in data] ref_end_index = [d[6] for d in data] # return torch.from_numpy(data2stack), torch.from_numpy(label2stack), grid_ind_stack, point_label, xyz, ref_st_index, ref_end_index # if ssl and multi frame: also add the start and end index of reference scan/frame and # confidence probability pseudo label elif len(data[0]) == 8: ref_st_index = [d[6] for d in data] ref_end_index = [d[7] for d in data] lcw2stack = np.stack([d[5] for d in data]).astype(np.float32) # return torch.from_numpy(data2stack), torch.from_numpy(label2stack), grid_ind_stack, point_label, xyz, ref_st_index, ref_end_index, lcw2stack # return torch.from_numpy(data2stack), torch.from_numpy(label2stack), grid_ind_stack, point_label, xyz return torch.from_numpy(data2stack), torch.from_numpy( label2stack), grid_ind_stack, point_label, xyz, ref_st_index, ref_end_index, lcw2stack def collate_fn_BEV_tta(data): data2stack = np.stack([da2[0] for da1 in data for da2 in da1]).astype(np.float32) label2stack = np.stack([da2[1] for da1 in data for da2 in da1]).astype(np.int) voxel_label = [] for da1 in data: for da2 in da1: voxel_label.append(da2[1]) #voxel_label.astype(np.int) grid_ind_stack = [] for da1 in data: for da2 in da1: grid_ind_stack.append(da2[2]) point_label = [] for da1 in data: for da2 in da1: point_label.append(da2[3]) xyz = [] for da1 in data: for da2 in da1: xyz.append(da2[4]) # index = [] # for da1 in data: # for da2 in da1: # index.append(da2[5]) ref_st_index = None ref_end_index = None lcw2stack = None # if multi frame but not ssl: also add the start and end index of reference scan/frame if len(data[0]) == 7: ref_st_index = [da2[5] for da1 in data for da2 in da1] ref_end_index = [da2[6] for da1 in data for da2 in da1] # return torch.from_numpy(data2stack), torch.from_numpy(label2stack), grid_ind_stack, point_label, xyz, ref_st_index, ref_end_index # if ssl and multi frame: also add the start and end index of reference scan/frame and # confidence probability pseudo label elif len(data[0]) == 8: ref_st_index = [da2[6] for da1 in data for da2 in da1] ref_end_index = [da2[7] for da1 in data for da2 in da1] lcw2stack = np.stack([da2[5] for da1 in data for da2 in da1]).astype(np.float32) # return xyz, voxel_label, grid_ind_stack, point_label, xyz, ref_st_index, ref_end_index, lcw2stack return torch.from_numpy(data2stack), torch.from_numpy( label2stack), grid_ind_stack, point_label, xyz, ref_st_index, ref_end_index, lcw2stack def collate_fn_BEV_test(data): data2stack = np.stack([d[0] for d in data]).astype(np.float32) label2stack = np.stack([d[1] for d in data]).astype(np.int) grid_ind_stack = [d[2] for d in data] point_label = [d[3] for d in data] xyz = [d[4] for d in data] index = [d[5] for d in data] return torch.from_numpy(data2stack), torch.from_numpy(label2stack), grid_ind_stack, point_label, xyz, index
29,622
40.372905
150
py
T-Concord3D
T-Concord3D-master/dataloader/augmentations.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------| # from __future__ import ( # division, # absolute_import, # with_statement, # print_function, # unicode_literals, # ) import random import numpy as np import torch #from pointnet2.data.data_utils import angle_axis # Utilis def angle_axis(angle, axis): # type: (float, np.ndarray) -> float r"""Returns a 4x4 rotation matrix that performs a rotation around axis by angle Parameters ---------- angle : float Angle to rotate by axis: np.ndarray Axis to rotate about Returns ------- torch.Tensor 3x3 rotation matrix """ u = axis / np.linalg.norm(axis) cosval, sinval = np.cos(angle), np.sin(angle) # yapf: disable cross_prod_mat = np.array([[0.0, -u[2], u[1]], [u[2], 0.0, -u[0]], [-u[1], u[0], 0.0]]) R = torch.from_numpy( cosval * np.eye(3) + sinval * cross_prod_mat + (1.0 - cosval) * np.outer(u, u) ) # yapf: enable return R.float() ##################################3 def RandomFlipX(points , v): assert 0 <= v <= 1 if np.random.random() < v: points[:,0] *= -1 return points def RandomFlipY(points , v): assert 0 <= v <= 1 if np.random.random() < v: points[:,1] *= -1 return points def RandomFlipZ(points , v): assert 0 <= v <= 1 if np.random.random() < v: points[:,2] *= -1 return points def ScaleX(pts,v): # (0 , 2) assert 0 <= v <= 0.5 scaler = np.random.uniform(low = 1-v, high = 1 + v) pts[:, 0 ] *= scaler return pts def ScaleY(pts,v): # (0 , 2) assert 0 <= v <= 0.5 scaler = np.random.uniform(low = 1-v, high = 1 + v) pts[:, 1 ] *= scaler return pts def ScaleZ(pts,v): # (0 , 2) assert 0 <= v <= 0.5 scaler = np.random.uniform(low = 1-v, high = 1 + v) pts[:, 2 ] *= scaler return pts def Resize(pts,v): assert 0 <= v <= 0.5 scaler = np.random.uniform(low = 1-v, high = 1 + v) pts[:, 0:3 ] *= scaler return pts def NonUniformScale(pts,v): # Resize in [0.5 , 1.5] assert 0 <= v <= 0.5 scaler = np.random.uniform( low = 1 - v, high = 1 + v, size = 3 ) pts[:, 0:3] *= torch.from_numpy(scaler).float() return pts def RotateX(points,v): # ( 0 , 2 * pi) assert 0 <= v <= 2 * np.pi if np.random.random() > 0.5: v *= -1 axis = np.array([1. , 0. , 0.]) rotation_angle = np.random.uniform() * v rotation_matrix = angle_axis(rotation_angle, axis) normals = points.size(1) > 3 if not normals: return points @ rotation_matrix.t() else: pc_xyz = points[:, 0:3] pc_normals = points[:, 3:] points[:, 0:3] = pc_xyz @ rotation_matrix.t() points[:, 3:] = pc_normals @ rotation_matrix.t() return points def RotateY(points,v): # ( 0 , 2 * pi) assert 0 <= v <= 2 * np.pi if np.random.random() > 0.5: v *= -1 axis = np.array([0. , 1. , 0.]) rotation_angle = np.random.uniform() * v rotation_matrix = angle_axis(rotation_angle, axis) normals = points.size(1) > 3 if not normals: return points @ rotation_matrix.t() else: pc_xyz = points[:, 0:3] pc_normals = points[:, 3:] points[:, 0:3] = pc_xyz @ rotation_matrix.t() points[:, 3:] = pc_normals @ rotation_matrix.t() return points def RotateZ(points,v): # ( 0 , 2 * pi) assert 0 <= v <= 2 * np.pi if np.random.random() > 0.5: v *= -1 axis = np.array([0. , 0. , 1.]) rotation_angle = np.random.uniform() * v rotation_matrix = angle_axis(rotation_angle, axis) normals = points.size(1) > 3 if not normals: return points @ rotation_matrix.t() else: pc_xyz = points[:, 0:3] pc_normals = points[:, 3:] points[:, 0:3] = pc_xyz @ rotation_matrix.t() points[:, 3:] = pc_normals @ rotation_matrix.t() return points def RandomAxisRotation(points,v): assert 0 <= v <= 2 * np.pi axis = np.random.randn(3) axis /= np.sqrt((axis**2).sum()) rotation_angle = np.random.uniform() * v rotation_matrix = angle_axis(rotation_angle, axis) normals = points.size(1) > 3 if not normals: return points @ rotation_matrix.t() else: pc_xyz = points[:, 0:3] pc_normals = points[:, 3:] points[:, 0:3] = pc_xyz @ rotation_matrix.t() points[:, 3:] = pc_normals @ rotation_matrix.t() return points def RotatePerturbation(points,v): assert 0 <= v <= 10 v = int(v) angle_sigma = 0.1 * v angle_clip = 0.1 * v n_idx = 50 * v angles = np.clip(angle_sigma * np.random.randn(3), -angle_clip, angle_clip) Rx = angle_axis(angles[0], np.array([1.0, 0.0, 0.0])) Ry = angle_axis(angles[1], np.array([0.0, 1.0, 0.0])) Rz = angle_axis(angles[2], np.array([0.0, 0.0, 1.0])) rotation_matrix = Rz @ Ry @ Rx center = torch.mean(points[:,0:3], dim = 0) idx = np.random.choice(points.size(0),n_idx) perturbation = points[idx, 0:3] - center points[idx, :3] += (perturbation @ rotation_matrix.t()) - perturbation normals = points.size(1) > 3 if normals: pc_normals = points[idx, 3:] points[idx, 3:] = pc_normals @ rotation_matrix.t() return points def Jitter(points,v): assert 0.0 <= v <= 10 v = int(v) sigma = 0.1 * v n_idx = 50 * v idx = np.random.choice(points.size(0),n_idx) jitter = sigma * (np.random.random([n_idx, 3]) - 0.5) points[idx, 0:3] += torch.from_numpy(jitter).float() return points def PointToNoise(points,v): assert 0 <= v <= 0.5 mask = np.random.random(points.size(0)) < v noise_idx = [idx for idx in range(len(mask)) if mask[idx] == True] pts_rand = 2 * (np.random.random([len(noise_idx), 3]) - 0.5) + np.mean(points[:,0:3].numpy(), axis= 0) points[noise_idx, 0:3] = torch.from_numpy(pts_rand).float() return points def UniformTranslate(points ,v): assert 0 <= v <= 1 translation = (2 * np.random.random() - 1 ) * v points[:, 0:3] += translation return points def NonUniformTranslate(points ,v): assert 0 <= v <= 1 translation = (2 * np.random.random(3) - 1 ) * v points[:, 0:3] += torch.from_numpy(translation).float() return points def RandomDropout(points,v): assert 0.3 <= v <= 0.875 dropout_rate = v drop = torch.rand(points.size(0)) < dropout_rate save_idx = np.random.randint(points.size(0)) points[drop] = points[save_idx] return points def RandomErase(points,v): assert 0 <= v <= 0.5 "v : the radius of erase ball" random_idx = np.random.randint(points.size(0)) mask = torch.sum((points[:,0:3] - points[random_idx,0:3]).pow(2), dim = 1) < v ** 2 points[mask] = points[random_idx] return points # # def DBSCAN(points,v): # "https://scikit-learn.org/stable/modules/generated/sklearn.cluster.DBSCAN.html" # assert 0 <= v <= 10 # from sklearn.cluster import DBSCAN # eps = 0.03 * v # min_samples = v # clustering = DBSCAN(eps = eps , min_samples = min_samples).fit(points[:,0:3].numpy()) # for label in set(clustering.labels_): # mask = (clustering.labels_ == label) # points[mask,0:3] = torch.mean(points[mask,0:3] , dim = 0) # # return points def ShearXY(points,v): assert 0 <= v <= 0.5 a , b = v * (2 * np.random.random(2) - 1) shear_matrix = np.array([[1, 0, 0], [0, 1, 0], [a, b, 1]]) shear_matrix = torch.from_numpy(shear_matrix).float() points[:,0:3] = points[:, 0:3] @ shear_matrix.t() return points def ShearYZ(points,v): assert 0 <= v <= 0.5 b , c = v * (2 * np.random.random(2) - 1) shear_matrix = np.array([[1, b, c], [0, 1, 0], [0, 0, 1]]) shear_matrix = torch.from_numpy(shear_matrix).float() points[:,0:3] = points[:, 0:3] @ shear_matrix.t() return points def ShearXZ(points,v): assert 0 <= v <= 0.5 a , c = v * (2 * np.random.random(2) - 1) shear_matrix = np.array([[1, 0, 0], [a, 1, c], [0, 0, 1]]) shear_matrix = torch.from_numpy(shear_matrix).float() points[:,0:3] = points[:, 0:3] @ shear_matrix.t() return points def GlobalAffine(points,v): assert 0 <= v <= 1 affine_matrix = torch.from_numpy(np.eye(3) + np.random.randn(3,3) * v).float() points[:,0:3] = points[:, 0:3] @ affine_matrix.t() return points def Identity(points , v): return points def augment_list(): # operations and their ranges l = ( (Identity , 0 ,10), (RandomFlipX, 0, 1), (RandomFlipY, 0, 1), (RandomFlipZ, 0, 1), (ScaleX, 0, 0.5), (ScaleY, 0, 0.5), (ScaleZ, 0, 0.5), (NonUniformScale, 0 , 0.5), (Resize , 0 , 0.5), (RotateX, 0, 2 * np.pi), (RotateY, 0, 2 * np.pi), (RotateZ, 0, 2 * np.pi), (RandomAxisRotation, 0, 2 * np.pi), (RotatePerturbation, 0, 10), (Jitter, 0 , 10), (UniformTranslate, 0 , 0.5), (NonUniformTranslate, 0 , 0.5), (RandomDropout, 0.3 , 0.875), (RandomErase, 0, 0.5), (PointToNoise, 0 , 0.5), (ShearXY, 0 , 0.5 ), (ShearYZ, 0 , 0.5 ), (ShearXZ, 0 , 0.5 ), (GlobalAffine , 0 , 0.15), ) return l # class RandAugment3D: # def __init__(self, n, m): # self.n = n # self.m = m # [0, 30] # self.augment_list = augment_list() # # def __call__(self, img): # ops = random.sample(self.augment_list, k=self.n) # for op in ops: # # val = (float(self.m) / 30) * float(maxval - minval) + minval # val = self.m # img = op(img, val) # # return img class RandAugment3D: def __init__(self, n=2, m=10): """ The number of augmentations = ? N : The number of augmentation choice M : magnitude of augmentation """ self.n = n self.m = m # [0, 10] self.augment_list = augment_list() self.epoch = 0 def __call__(self, pc): assert 0 <= self.m <= 10 if pc.dim() == 3: bsize = pc.size()[0] for i in range(bsize): points = pc[i,:,:] ops = random.choices(self.augment_list, k=self.n) for op, minval, maxval in ops: val = float(self.m) points = op(points, val) elif pc.dim() == 2: points = pc ops = random.choices(self.augment_list, k=self.n) for op, minval, maxval in ops: val = (float(self.m) / 10) * float(maxval - minval) + minval points = op(points, val) return pc def UpdateNM(self,increase=True): N_tmp, M_tmp = self.n, self.m if increase: if np.random.random() > 0.5: self.n += 1 elif self.m < 10: self.m += 1 print("\n Increase N,M from ({},{}) to ({} ,{}) \n".format(N_tmp, M_tmp, self.n, self.m)) elif increase == False: if np.random.random() > 0.5 and self.n > 1: self.n -= 1 elif self.m > 1: self.m -= 1 print("\n Decrease N,M from ({},{}) to ({} ,{}) \n".format(N_tmp, M_tmp, self.n, self.m))
11,680
27.079327
106
py
T-Concord3D
T-Concord3D-master/dataloader/pc_dataset.py
# -*- coding:utf-8 -*- # author: Xinge # @file: pc_dataset.py import glob import os import pickle from os.path import exists import numpy as np import yaml from torch.utils import data REGISTERED_PC_DATASET_CLASSES = {} # past and future frames global place holders past = 0 future = 0 T_past = 0 T_future = 0 ssl = False rgb = False def register_dataset(cls, name=None): global REGISTERED_PC_DATASET_CLASSES if name is None: name = cls.__name__ assert name not in REGISTERED_PC_DATASET_CLASSES, f"exist class: {REGISTERED_PC_DATASET_CLASSES}" REGISTERED_PC_DATASET_CLASSES[name] = cls return cls def get_pc_model_class(name): global REGISTERED_PC_DATASET_CLASSES assert name in REGISTERED_PC_DATASET_CLASSES, f"available class: {REGISTERED_PC_DATASET_CLASSES}" return REGISTERED_PC_DATASET_CLASSES[name] @register_dataset class SemKITTI_demo(data.Dataset): def __init__(self, data_path, imageset='demo', return_ref=True, label_mapping="semantic-wod.yaml", demo_label_path=None): with open(label_mapping, 'r') as stream: semkittiyaml = yaml.safe_load(stream) self.learning_map = semkittiyaml['learning_map'] self.imageset = imageset self.return_ref = return_ref self.im_idx = [] self.im_idx += absoluteFilePaths(data_path) self.label_idx = [] if self.imageset == 'val': print(demo_label_path) self.label_idx += absoluteFilePaths(demo_label_path) def __len__(self): 'Denotes the total number of samples' return len(self.im_idx) def __getitem__(self, index): raw_data = np.fromfile(self.im_idx[index], dtype=np.float32).reshape((-1, 4)) if self.imageset == 'demo': annotated_data = np.expand_dims(np.zeros_like(raw_data[:, 0], dtype=int), axis=1) elif self.imageset == 'val': annotated_data = np.fromfile(self.label_idx[index], dtype=np.uint32).reshape((-1, 1)) annotated_data = annotated_data & 0xFFFF # delete high 16 digits binary annotated_data = np.vectorize(self.learning_map.__getitem__)(annotated_data) data_tuple = (raw_data[:, :3], annotated_data.astype(np.uint8)) if self.return_ref: data_tuple += (raw_data[:, 3],) return data_tuple @register_dataset class SemKITTI_sk(data.Dataset): def __init__(self, data_path, imageset='train', return_ref=False, label_mapping="semantic-wod.yaml", nusc=None, ssl_data_path=None): self.return_ref = return_ref with open(label_mapping, 'r') as stream: semkittiyaml = yaml.safe_load(stream) self.learning_map = semkittiyaml['learning_map'] self.imageset = imageset if imageset == 'train': split = semkittiyaml['split']['train'] elif imageset == 'val': split = semkittiyaml['split']['valid'] elif imageset == 'test': split = semkittiyaml['split']['test'] else: raise Exception('Split must be train/val/test') global past, future, ssl, T_past, T_fture self.past = past self.future = future self.T_past = T_past self.T_future = T_future self.im_idx = [] for i_folder in split: self.im_idx += absoluteFilePaths('/'.join([data_path, str(i_folder).zfill(2), 'velodyne'])) def __len__(self): 'Denotes the total number of samples' return len(self.im_idx) def __getitem__(self, index): raw_data = np.fromfile(self.im_idx[index], dtype=np.float32).reshape((-1, 4)) ''' if self.imageset == 'test': annotated_data = np.expand_dims(np.zeros_like(raw_data[:, 0], dtype=int), axis=1) else: annotated_data = np.fromfile(self.im_idx[index].replace('velodyne', 'labels')[:-3] + 'label', dtype=np.uint32).reshape((-1, 1)) annotated_data = annotated_data & 0xFFFF # delete high 16 digits binary annotated_data = np.vectorize(self.learning_map.__getitem__)(annotated_data) data_tuple = (raw_data[:, :3], annotated_data.astype(np.uint8)) if self.return_ref: data_tuple += (raw_data[:, 3],) return data_tuple ''' origin_len = len(raw_data) if self.imageset == 'test': annotated_data = np.expand_dims(np.zeros_like(raw_data[:, 0], dtype=int), axis=1) else: # x = self.im_idx[index].replace('velodyne', f"predictions_{self.T_past}_{self.T_future}")[:-3] + 'label' if ssl and exists(self.im_idx[index].replace('velodyne', f"predictions_{self.T_past}_{self.T_future}")[ :-3] + 'label'): annotated_data = np.fromfile( self.im_idx[index].replace('velodyne', f"predictions_{self.T_past}_{self.T_future}")[ :-3] + 'label', dtype=np.int32).reshape((-1, 1)) else: annotated_data = np.fromfile(self.im_idx[index].replace('velodyne', 'labels')[:-3] + 'label', dtype=np.int32).reshape((-1, 1)) annotated_data = annotated_data & 0xFFFF # delete high 16 digits binary # annotated_data = np.vectorize(self.learning_map.__getitem__)(annotated_data) if ssl and exists(self.im_idx[index].replace('velodyne', f"probability_{self.T_past}_{self.T_future}")[ :-3] + 'label'): lcw = np.fromfile(self.im_idx[index].replace('velodyne', f"probability_{self.T_past}_{self.T_future}")[ :-3] + 'label', dtype=np.float32).reshape((-1, 1)) # TODO: check casting lcw = (lcw * 100).astype(np.int32) elif ssl: # in case of GT label give weight = 1.0 per label lcw = np.expand_dims(np.ones_like(raw_data[:, 0], dtype=np.float32), axis=1) # TODO: check casting lcw = (lcw * 100).astype(np.int32) number_idx = int(self.im_idx[index][-10:-4]) dir_idx = int(self.im_idx[index][-22:-20]) past_frame_len = 0 future_frame_len = 0 annotated_data = np.vectorize(self.learning_map.__getitem__)(annotated_data) data_tuple = (raw_data[:, :3], annotated_data.astype(np.uint8)) if self.return_ref and ssl: # np.save('pcl.npy', raw_data[:, :3]) # np.save('label.npy', annotated_data) # TODO: masking below 0.8 confidence # lcw_mask = lcw < 80 # lcw[lcw_mask] = 0 data_tuple += (raw_data[:, 3], lcw, future_frame_len, origin_len) # origin_len is used to indicate the length of target-scan and lcw elif self.return_ref: data_tuple += ( raw_data[:, 3], future_frame_len, origin_len) # origin_len is used to indicate the length of target-scan return data_tuple @register_dataset class SemKITTI_nusc(data.Dataset): def __init__(self, data_path, imageset='train', return_ref=False, label_mapping="nuscenes.yaml", nusc=None): self.return_ref = return_ref with open(imageset, 'rb') as f: data = pickle.load(f) with open(label_mapping, 'r') as stream: nuscenesyaml = yaml.safe_load(stream) self.learning_map = nuscenesyaml['learning_map'] self.nusc_infos = data['infos'] self.data_path = data_path self.nusc = nusc def __len__(self): 'Denotes the total number of samples' return len(self.nusc_infos) def __getitem__(self, index): info = self.nusc_infos[index] lidar_path = info['lidar_path'][16:] lidar_sd_token = self.nusc.get('sample', info['token'])['data']['LIDAR_TOP'] lidarseg_labels_filename = os.path.join(self.nusc.dataroot, self.nusc.get('lidarseg', lidar_sd_token)['filename']) points_label = np.fromfile(lidarseg_labels_filename, dtype=np.uint8).reshape([-1, 1]) points_label = np.vectorize(self.learning_map.__getitem__)(points_label) points = np.fromfile(os.path.join(self.data_path, lidar_path), dtype=np.float32, count=-1).reshape([-1, 5]) data_tuple = (points[:, :3], points_label.astype(np.uint8)) if self.return_ref: data_tuple += (points[:, 3],) return data_tuple def absoluteFilePaths(directory): for dirpath, _, filenames in os.walk(directory): filenames.sort() for f in filenames: yield os.path.abspath(os.path.join(dirpath, f)) def SemKITTI2train(label): if isinstance(label, list): return [SemKITTI2train_single(a) for a in label] else: return SemKITTI2train_single(label) def SemKITTI2train_single(label): remove_ind = label == 0 label -= 1 label[remove_ind] = 255 return label from os.path import join def transform_pcl_scan(points, pose0, pose): # pose = poses[0][idx] hpoints = np.hstack((points[:, :3], np.ones_like(points[:, :1]))) # new_points = hpoints.dot(pose.T) new_points = np.sum(np.expand_dims(hpoints, 2) * pose.T, axis=1) new_points = new_points[:, :3] new_coords = new_points - pose0[:3, 3] # new_coords = new_coords.dot(pose0[:3, :3]) new_coords = np.sum(np.expand_dims(new_coords, 2) * pose0[:3, :3], axis=1) new_coords = np.hstack((new_coords, points[:, 3:])) return new_coords def fuse_multiscan(ref_raw_data, ref_annotated_data, ref_lcw, transformed_data, transformed_annotated_data, transformed_lcw, source, ssl): lcw = None if (source != 1) and (source != -1): print(f"Error data source {source} not Implemented") return 0 if source == -1: # past frame raw_data = np.concatenate((transformed_data, ref_raw_data), 0) annotated_data = np.concatenate((transformed_annotated_data, ref_annotated_data), 0) if ssl: lcw = np.concatenate((transformed_lcw, ref_lcw), 0) if source == 1: # future frame raw_data = np.concatenate((ref_raw_data, transformed_data,), 0) annotated_data = np.concatenate((ref_annotated_data, transformed_annotated_data), 0) if ssl: lcw = np.concatenate((ref_lcw, transformed_lcw), 0) return raw_data, annotated_data, lcw def get_combined_data(raw_data, annotated_data, lcw, learning_map, return_ref, origin_len, preceding_frame_len, ssl): #print(np.unique(annotated_data)) annotated_data = np.vectorize(learning_map.__getitem__)(annotated_data) data_tuple = (raw_data[:, :3], annotated_data.astype(np.uint8)) if return_ref and ssl: # np.save('pcl.npy', raw_data[:, :3]) # np.save('label.npy', annotated_data) # TODO: masking below 0.8 confidence # lcw_mask = lcw < 80 # lcw[lcw_mask] = 0 # origin_len is used to indicate the length of target-scan and lcw data_tuple += (raw_data[:, 3], lcw, preceding_frame_len, origin_len) elif return_ref: # origin_len is used to indicate the length of target-scan data_tuple += (raw_data[:, 3], preceding_frame_len, origin_len) return data_tuple @register_dataset class SemKITTI_sk_multiscan(data.Dataset): def __init__(self, data_path, imageset='train', return_ref=False, label_mapping="semantic-kitti-multiscan.yaml", train_hypers=None, wod=None, ssl_data_path=None): global past, future, ssl, T_past, T_fture with open(label_mapping, 'r') as stream: semkittiyaml = yaml.safe_load(stream) self.return_ref = return_ref self.learning_map = semkittiyaml['learning_map'] self.imageset = imageset self.data_path = data_path self.past = train_hypers['past'] self.future = train_hypers['future'] self.T_past = train_hypers['T_past'] self.T_future = train_hypers['T_future'] self.ssl = train_hypers['ssl'] self.im_idx = [] self.calibrations = [] # self.times = [] self.poses = [] if imageset == 'train': self.split = semkittiyaml['split']['train'] if self.ssl and (ssl_data_path is not None): self.split += semkittiyaml['split']['pseudo'] elif imageset == 'val': self.split = semkittiyaml['split']['valid'] elif imageset == 'test': self.split = semkittiyaml['split']['test'] elif imageset == 'pseudo': self.split = semkittiyaml['split']['pseudo'] else: raise Exception(f'{imageset}: Split must be train/val/test/pseudo') if self.past or self.future: self.load_calib_poses() for i_folder in self.split: self.im_idx += absoluteFilePaths('/'.join([data_path, str(i_folder).zfill(2), 'velodyne'])) def __len__(self): 'Denotes the total number of samples' return len(self.im_idx) def load_calib_poses(self): """ load calib poses and times. """ ########### # Load data ########### self.calibrations = [] # self.times = [] self.poses = {} # [] for seq in self.split: seq_folder = join(self.data_path, str(seq).zfill(2)) # Read Calib self.calibrations.append(self.parse_calibration(join(seq_folder, "calib.txt"))) # Read times # self.times.append(np.loadtxt(join(seq_folder, 'times.txt'), dtype=np.float32)) # Read poses poses_f64 = self.parse_poses(join(seq_folder, 'poses.txt'), self.calibrations[-1]) # self.poses.append([pose.astype(np.float32) for pose in poses_f64]) self.poses[seq] = [pose.astype(np.float32) for pose in poses_f64] def parse_calibration(self, filename): """ read calibration file with given filename Returns ------- dict Calibration matrices as 4x4 numpy arrays. """ calib = {} calib_file = open(filename) # print(filename) for line in calib_file: key, content = line.strip().split(":") values = [float(v) for v in content.strip().split()] pose = np.zeros((4, 4)) pose[0, 0:4] = values[0:4] pose[1, 0:4] = values[4:8] pose[2, 0:4] = values[8:12] pose[3, 3] = 1.0 calib[key] = pose calib_file.close() return calib def parse_poses(self, filename, calibration): """ read poses file with per-scan poses from given filename Returns ------- list list of poses as 4x4 numpy arrays. """ file = open(filename) # print(filename) poses = [] Tr = calibration["Tr"] Tr_inv = np.linalg.inv(Tr) for line in file: values = [float(v) for v in line.strip().split()] pose = np.zeros((4, 4)) pose[0, 0:4] = values[0:4] pose[1, 0:4] = values[4:8] pose[2, 0:4] = values[8:12] pose[3, 3] = 1.0 poses.append(np.matmul(Tr_inv, np.matmul(pose, Tr))) return poses def get_semantickitti_data(self, newpath, time_frame_idx): raw_data = np.fromfile(newpath, dtype=np.float32).reshape((-1, 4)) lcw = None if self.imageset == 'test' or self.imageset == 'pseudo': annotated_data = np.expand_dims(np.zeros_like(raw_data[:, 0], dtype=int), axis=1) else: if self.ssl and exists(newpath.replace('velodyne', f"predictions_{self.T_past}_{self.T_future}")[:-3] + 'label'): annotated_data = np.fromfile( newpath.replace('velodyne', f"predictions_{self.T_past}_{self.T_future}")[:-3] + 'label', dtype=np.int64).reshape((-1, 1)) else: annotated_data = np.fromfile(newpath.replace('velodyne', 'labels')[:-3] + 'label', dtype=np.int32).reshape((-1, 1)) annotated_data = annotated_data & 0xFFFF # delete high 16 digits binary # if np.sum(np.unique(annotated_data == 18)) > 0: # print(newpath) if self.ssl and exists(newpath.replace('velodyne', f"probability_{self.T_past}_{self.T_future}")[ :-3] + 'label'): lcw = np.fromfile( newpath.replace('velodyne', f"probability_{self.T_past}_{self.T_future}")[ :-3] + 'label', dtype=np.float32).reshape((-1, 1)) # TODO: check casting lcw = (lcw * 100).astype(np.int32) elif self.ssl: # in case of GT label give weight = 1.0 per label lcw = np.expand_dims(np.ones_like(raw_data[:, 0], dtype=np.float32), axis=1) # TODO: check casting lcw = (lcw * 100).astype(np.int32) return raw_data, annotated_data, len(raw_data), lcw def __getitem__(self, index): # reference scan reference_file = self.im_idx[index] raw_data, annotated_data, data_len, lcw = self.get_semantickitti_data(reference_file, 0) origin_len = data_len number_idx = int(self.im_idx[index][-10:-4]) # dir_idx = int(self.im_idx[index][-22:-20]) dir_idx = self.im_idx[index].split('/')[-3] # past scan past_frame_len = 0 # TODO: added the future frame availability check if self.past and ((number_idx - self.past) >= 0) and ((number_idx + self.past) < len(self.poses[dir_idx])): # extract the poss of the reference frame pose0 = self.poses[dir_idx][number_idx] for fuse_idx in range(self.past): # TODO: past frames frame_ind = fuse_idx + 1 pose = self.poses[dir_idx][number_idx - frame_ind] past_file = self.im_idx[index][:-10] + str(number_idx - frame_ind).zfill(6) + self.im_idx[index][-4:] past_raw_data, past_annotated_data, past_data_len, past_lcw = self.get_semantickitti_data(past_file, -frame_ind) past_raw_data = transform_pcl_scan(past_raw_data, pose0, pose) # past frames if past_data_len != 0: raw_data, annotated_data, lcw = fuse_multiscan(raw_data, annotated_data, lcw, past_raw_data, past_annotated_data, past_lcw, -1, self.ssl) # count number of past frame points past_frame_len += past_data_len # future scan future_frame_len = 0 # TODO: added the future frame availability check if self.future and ((number_idx - self.future) >= 0) and ( (number_idx + self.future) < len(self.poses[dir_idx])): # extract the poss of the reference frame pose0 = self.poses[dir_idx][number_idx] for fuse_idx in range(self.future): # TODO: future frame frame_ind = fuse_idx + 1 future_pose = self.poses[dir_idx][number_idx + frame_ind] future_file = self.im_idx[index][:-10] + str(number_idx + frame_ind).zfill(6) + self.im_idx[index][-4:] future_raw_data, future_annotated_data, future_data_len, future_lcw = self.get_semantickitti_data( future_file, frame_ind) future_raw_data = transform_pcl_scan(future_raw_data, pose0, future_pose) # TODO: check correctness (future frame) if future_data_len != 0: raw_data, annotated_data, lcw = fuse_multiscan(raw_data, annotated_data, lcw, future_raw_data, future_annotated_data, future_lcw, 1, self.ssl) # count number of future frame points future_frame_len += future_data_len # extract compiled data_tuple data_tuple = get_combined_data(raw_data, annotated_data, lcw, self.learning_map, self.return_ref, origin_len, past_frame_len, self.ssl) # # TODO: added the future frame availability check return data_tuple # WOD ------------------------------------------------------------- # def fuse_multi_scan(points, pose0, pose): # # pose = poses[0][idx] # # hpoints = np.hstack((points[:, :3], np.ones_like(points[:, :1]))) # # new_points = hpoints.dot(pose.T) # new_points = np.sum(np.expand_dims(hpoints, 2) * pose.T, axis=1) # # new_points = new_points[:, :3] # new_coords = new_points - pose0[:3, 3] # # new_coords = new_coords.dot(pose0[:3, :3]) # new_coords = np.sum(np.expand_dims(new_coords, 2) * pose0[:3, :3], axis=1) # new_coords = np.hstack((new_coords, points[:, 3:])) # # return new_coords @register_dataset class WOD_multiscan(data.Dataset): def __init__(self, data_path, imageset='train', return_ref=False, label_mapping="wod-multiscan_labelled.yaml", train_hypers=None, wod=None, ssl_data_path=None): global past, future, ssl, T_past, T_fture, rgb self.return_ref = return_ref with open(label_mapping, 'r') as stream: wodyaml = yaml.safe_load(stream) self.learning_map = wodyaml['learning_map'] self.imageset = imageset self.data_path = data_path self.past = train_hypers['past'] self.future = train_hypers['future'] self.T_past = train_hypers['T_past'] self.T_future = train_hypers['T_future'] self.rgb = train_hypers['rgb'] self.ssl = train_hypers['ssl'] # self.use_time = train_hypers['time'] # Use time instead of intensity # self.UDA = train_hypers['uda'] self.im_idx = [] self.calibrations = [] # self.times = [] self.poses = {} if imageset == 'train': self.split = wodyaml['split']['train'] # self.sensor_zpose = train_hypers["S_sensor_zpose"] if self.ssl and (ssl_data_path is not None): self.split += wodyaml['split']['pseudo'] elif imageset == 'val': self.split = wodyaml['split']['valid'] # self.sensor_zpose = train_hypers["S_sensor_zpose"] elif imageset == 'test': self.split = wodyaml['split']['test'] # self.sensor_zpose = train_hypers["S_sensor_zpose"] elif imageset == 'pseudo': self.split = wodyaml['split']['pseudo'] # self.sensor_zpose = train_hypers["T_sensor_zpose"] else: raise Exception(f'{imageset}: Split must be train/val/test/pseudo') # self.split = sorted(os.listdir(self.data_path)) # self.training_len = len(self.split) # self.ssl_data_path = ssl_data_path # '/mnt/beegfs/gpu/argoverse-tracking-all-training/WOD/processed/Unlabeled/testing' # xx = self.data_path.split("/")[-1] # if ssl and self.data_path.split("/")[-1] == "training" and self.ssl_data_path: # self.training_len = len(sorted(os.listdir(self.data_path))) # self.split = sorted(os.listdir(self.data_path)) + sorted(os.listdir(self.ssl_data_path)) # print(len(self.split)) # TODO: remove after search experiment # self.split = self.split[150:] if self.past or self.future: self.load_calib_poses() # for c, i_folder in enumerate(self.split): # if ssl and (self.data_path.split("/")[-1] == "training") and ( # c >= self.training_len): # 789 number of training folders # self.im_idx += absoluteFilePaths('/'.join([self.ssl_data_path, str(i_folder), 'lidar'])) # else: # self.im_idx += absoluteFilePaths('/'.join([self.data_path, str(i_folder), 'lidar'])) for c, i_folder in enumerate(self.split): self.im_idx += absoluteFilePaths('/'.join([self.data_path, str(i_folder), 'lidar'])) def __len__(self): 'Denotes the total number of samples' return len(self.im_idx) def load_calib_poses(self): """ load calib poses and times. """ ########### # Load data ########### self.calibrations = [] # self.times = [] self.poses = {} # [] for k, seq in enumerate(self.split): # range(0, 22): # if ssl and (self.data_path.split("/")[-1] == "training") and ( # k >= self.training_len): # 789 number of training folders # seq_folder = join(self.ssl_data_path, str(seq)) # else: seq_folder = join(self.data_path, str(seq)) # Read poses poses_f64 = self.parse_poses(seq_folder, k) # self.poses.append([pose.astype(np.float32) for pose in poses_f64]) self.poses[seq] = [pose.astype(np.float32) for pose in poses_f64] def parse_poses(self, seq, k): """ read poses file with per-scan poses from given filename Returns ------- list list of poses as 4x4 numpy arrays. """ filename = sorted(glob.glob(os.path.join(seq, "poses", "*.npy"))) poses = [] for file in filename: pose = np.load(file) poses.append(pose) return poses def get_wod_data(self, newpath, time_frame_idx): raw_data = np.load(newpath) # if self.use_time: # raw_data[:, 3] = np.ones_like(raw_data[:, 3]) * time_frame_idx # if self.UDA: # raw_data[:, 2] += self.sensor_zpose # elevate the point cloud two meters up to align with WOD # TODO: check if the colors are encoded correctly instead of the lidar intensity if self.rgb: # load rgb colors for each points raw_rgb = np.load(newpath.replace('lidar', 'colors')[ :-3] + 'npy') # convert rgb into gray scale [0, 255] raw_gray = 0.2989 * raw_rgb[:, 0] + 0.5870 * raw_rgb[:, 1] + 0.1140 * raw_rgb[:, 2] # mask (0) ignored point colors (originally not provided on wod rear-cameras) -> rgb:[1,1,1] or gray:[ # 0.99990]) gray_mask = raw_gray > 1 # < 1 #0.9998999999999999 # assign 0 to the place we want to mask raw_gray[gray_mask] = -1 # replace intensity with gray scale camera image/frame color raw_data[:, 3] = raw_gray # raw_data[:,4] = gray_mask * 1 lcw = None origin_len = len(raw_data) if self.imageset == 'test' or self.imageset == 'pseudo': annotated_data = np.expand_dims(np.zeros_like(raw_data[:, 0]), axis=1).reshape((-1, 1)) else: # x = self.im_idx[index].replace('lidar', f"predictions_{self.T_past}_{self.T_future}")[:-3] + 'label' if self.ssl and exists(newpath.replace('lidar', f"predictions_{self.T_past}_{self.T_future}")[ :-3] + 'npy'): annotated_data = np.load( newpath.replace('lidar', f"predictions_{self.T_past}_{self.T_future}")[ :-3] + 'npy').reshape((-1, 1)) else: # print(self.im_idx[index].replace('lidar', 'labels')[:-3] + 'npy') annotated_data = np.load(newpath.replace('lidar', 'labels')[:-3] + 'npy', allow_pickle=True) if len(annotated_data.shape) == 2: if annotated_data.shape[1] == 2: annotated_data = annotated_data[:, 1] # Reshape the label/annotation to vector. annotated_data = annotated_data.reshape((-1, 1)) annotated_data = annotated_data & 0xFFFF # delete high 16 digits binary if self.ssl and exists(newpath.replace('lidar', f"probability_{self.T_past}_{self.T_future}")[ :-3] + 'npy'): lcw = np.load(newpath.replace('lidar', f"probability_{self.T_past}_{self.T_future}")[ :-3] + 'npy').reshape((-1, 1)) # TODO: check casting lcw = (lcw * 100).astype(np.int32) elif self.ssl: # in case of GT label give weight = 1.0 per label lcw = np.expand_dims(np.ones_like(raw_data[:, 0]), axis=1) # TODO: check casting lcw = (lcw * 100).astype(np.int32) return raw_data, annotated_data, len(raw_data), lcw def __getitem__(self, index): # reference scan reference_file = self.im_idx[index] raw_data, annotated_data, data_len, lcw = self.get_wod_data(reference_file, 0) origin_len = data_len number_idx = int(self.im_idx[index][-10:-4]) # dir_idx = int(self.im_idx[index][-22:-20]) dir_idx = self.im_idx[index].split('/')[-3] # past scan past_frame_len = 0 # TODO: added the future frame availability check if self.past and ((number_idx - self.past) >= 0) and ((number_idx + self.past) < len(self.poses[dir_idx])): # extract the poss of the reference frame pose0 = self.poses[dir_idx][number_idx] for fuse_idx in range(self.past): # TODO: past frames frame_ind = fuse_idx + 1 pose = self.poses[dir_idx][number_idx - frame_ind] past_file = self.im_idx[index][:-10] + str(number_idx - frame_ind).zfill(6) + self.im_idx[index][-4:] past_raw_data, past_annotated_data, past_data_len, past_lcw = self.get_wod_data(past_file, -frame_ind) # transform the past frame into reference frame coordinate system past_raw_data = transform_pcl_scan(past_raw_data, pose0, pose) # past frames if past_data_len != 0: raw_data, annotated_data, lcw = fuse_multiscan(raw_data, annotated_data, lcw, past_raw_data, past_annotated_data, past_lcw, -1, self.ssl) # count number of past frame points past_frame_len += past_data_len # future scan future_frame_len = 0 # TODO: added the future frame availability check if self.future and ((number_idx - self.future) >= 0) and ( (number_idx + self.future) < len(self.poses[dir_idx])): # extract the poss of the reference frame pose0 = self.poses[dir_idx][number_idx] for fuse_idx in range(self.future): # TODO: future frame frame_ind = fuse_idx + 1 future_pose = self.poses[dir_idx][number_idx + frame_ind] future_file = self.im_idx[index][:-10] + str(number_idx + frame_ind).zfill(6) + self.im_idx[index][-4:] future_raw_data, future_annotated_data, future_data_len, future_lcw = self.get_wod_data(future_file, frame_ind) # transform the future frame into reference frame coordinate system future_raw_data = transform_pcl_scan(future_raw_data, pose0, future_pose) # TODO: check correctness (future frame) if future_data_len != 0: raw_data, annotated_data, lcw = fuse_multiscan(raw_data, annotated_data, lcw, future_raw_data, future_annotated_data, future_lcw, 1, self.ssl) # count number of future frame points future_frame_len += future_data_len # extract compiled data_tuple data_tuple = get_combined_data(raw_data, annotated_data, lcw, self.learning_map, self.return_ref, origin_len, past_frame_len, self.ssl) return data_tuple # load label class info def get_label_name(label_mapping): with open(label_mapping, 'r') as stream: config_yaml = yaml.safe_load(stream) class_label_name = dict() for i in sorted(list(config_yaml['learning_map'].keys()))[::-1]: class_label_name[config_yaml['learning_map'][i]] = config_yaml['labels'][i] return class_label_name def get_label_inv_name(label_inv_mapping): with open(label_inv_mapping, 'r') as stream: config_yaml = yaml.safe_load(stream) # label_inv_name = dict() label_inv_name = config_yaml['learning_map_inv'] return label_inv_name def get_nuScenes_label_name(label_mapping): with open(label_mapping, 'r') as stream: nuScenesyaml = yaml.safe_load(stream) nuScenes_label_name = dict() for i in sorted(list(nuScenesyaml['learning_map'].keys()))[::-1]: val_ = nuScenesyaml['learning_map'][i] nuScenes_label_name[val_] = nuScenesyaml['labels_16'][val_] return nuScenes_label_name def update_config(configs): global past, future, T_past, T_future, ssl, rgb train_hypers = configs['train_params'] past = train_hypers['past'] future = train_hypers['future'] T_past = train_hypers['T_past'] T_future = train_hypers['T_future'] ssl = train_hypers['ssl'] rgb = train_hypers['rgb']
34,323
40.354217
130
py
T-Concord3D
T-Concord3D-master/dataloader/dataset_nuscenes.py
# -*- coding:utf-8 -*- # author: Xinge # @file: dataset_nuscenes.py import numpy as np import torch import numba as nb from torch.utils import data from dataloader.dataset_semantickitti import register_dataset def cart2polar(input_xyz): rho = np.sqrt(input_xyz[:, 0] ** 2 + input_xyz[:, 1] ** 2) phi = np.arctan2(input_xyz[:, 1], input_xyz[:, 0]) return np.stack((rho, phi, input_xyz[:, 2]), axis=1) def polar2cat(input_xyz_polar): x = input_xyz_polar[0] * np.cos(input_xyz_polar[1]) y = input_xyz_polar[0] * np.sin(input_xyz_polar[1]) return np.stack((x, y, input_xyz_polar[2]), axis=0) @register_dataset class cylinder_dataset_nuscenes(data.Dataset): def __init__(self, in_dataset, grid_size, rotate_aug=False, flip_aug=False, ignore_label=0, return_test=False, fixed_volume_space=False, max_volume_space=[50, np.pi, 3], min_volume_space=[0, -np.pi, -5], scale_aug=False, transform_aug=False, trans_std=[0.1, 0.1, 0.1], min_rad=-np.pi / 4, max_rad=np.pi / 4): 'Initialization' self.point_cloud_dataset = in_dataset self.grid_size = np.asarray(grid_size) self.rotate_aug = rotate_aug self.flip_aug = flip_aug self.scale_aug = scale_aug self.ignore_label = ignore_label self.return_test = return_test self.fixed_volume_space = fixed_volume_space self.max_volume_space = max_volume_space self.min_volume_space = min_volume_space self.transform = transform_aug self.trans_std = trans_std self.noise_rotation = np.random.uniform(min_rad, max_rad) def __len__(self): return len(self.point_cloud_dataset) def __getitem__(self, index): data = self.point_cloud_dataset[index] if len(data) == 2: xyz, labels = data elif len(data) == 3: xyz, labels, sig = data if len(sig.shape) == 2: sig = np.squeeze(sig) else: raise Exception('Return invalid data tuple') # random data augmentation by rotation if self.rotate_aug: rotate_rad = np.deg2rad(np.random.random() * 360) - np.pi c, s = np.cos(rotate_rad), np.sin(rotate_rad) j = np.matrix([[c, s], [-s, c]]) xyz[:, :2] = np.dot(xyz[:, :2], j) # random data augmentation by flip x , y or x+y if self.flip_aug: flip_type = np.random.choice(4, 1) if flip_type == 1: xyz[:, 0] = -xyz[:, 0] elif flip_type == 2: xyz[:, 1] = -xyz[:, 1] elif flip_type == 3: xyz[:, :2] = -xyz[:, :2] if self.scale_aug: noise_scale = np.random.uniform(0.95, 1.05) xyz[:, 0] = noise_scale * xyz[:, 0] xyz[:, 1] = noise_scale * xyz[:, 1] # convert coordinate into polar coordinates if self.transform: noise_translate = np.array([np.random.normal(0, self.trans_std[0], 1), np.random.normal(0, self.trans_std[1], 1), np.random.normal(0, self.trans_std[2], 1)]).T xyz[:, 0:3] += noise_translate xyz_pol = cart2polar(xyz) max_bound_r = np.percentile(xyz_pol[:, 0], 100, axis=0) min_bound_r = np.percentile(xyz_pol[:, 0], 0, axis=0) max_bound = np.max(xyz_pol[:, 1:], axis=0) min_bound = np.min(xyz_pol[:, 1:], axis=0) max_bound = np.concatenate(([max_bound_r], max_bound)) min_bound = np.concatenate(([min_bound_r], min_bound)) if self.fixed_volume_space: max_bound = np.asarray(self.max_volume_space) min_bound = np.asarray(self.min_volume_space) # get grid index crop_range = max_bound - min_bound cur_grid_size = self.grid_size intervals = crop_range / (cur_grid_size - 1) if (intervals == 0).any(): print("Zero interval!") grid_ind = (np.floor((np.clip(xyz_pol, min_bound, max_bound) - min_bound) / intervals)).astype(np.int) voxel_position = np.zeros(self.grid_size, dtype=np.float32) dim_array = np.ones(len(self.grid_size) + 1, int) dim_array[0] = -1 voxel_position = np.indices(self.grid_size) * intervals.reshape(dim_array) + min_bound.reshape(dim_array) voxel_position = polar2cat(voxel_position) # process labels processed_label = np.ones(self.grid_size, dtype=np.uint8) * self.ignore_label label_voxel_pair = np.concatenate([grid_ind, labels], axis=1) label_voxel_pair = label_voxel_pair[np.lexsort((grid_ind[:, 0], grid_ind[:, 1], grid_ind[:, 2])), :] processed_label = nb_process_label(np.copy(processed_label), label_voxel_pair) data_tuple = (voxel_position, processed_label) # center data on each voxel for PTnet voxel_centers = (grid_ind.astype(np.float32) + 0.5) * intervals + min_bound return_xyz = xyz_pol - voxel_centers return_xyz = np.concatenate((return_xyz, xyz_pol, xyz[:, :2]), axis=1) if len(data) == 2: return_fea = return_xyz elif len(data) == 3: return_fea = np.concatenate((return_xyz, sig[..., np.newaxis]), axis=1) if self.return_test: data_tuple += (grid_ind, labels, return_fea, index) else: data_tuple += (grid_ind, labels, return_fea) return data_tuple @nb.jit('u1[:,:,:](u1[:,:,:],i8[:,:])', nopython=True, cache=True, parallel=False) def nb_process_label(processed_label, sorted_label_voxel_pair): label_size = 256 counter = np.zeros((label_size,), dtype=np.uint16) counter[sorted_label_voxel_pair[0, 3]] = 1 cur_sear_ind = sorted_label_voxel_pair[0, :3] for i in range(1, sorted_label_voxel_pair.shape[0]): cur_ind = sorted_label_voxel_pair[i, :3] if not np.all(np.equal(cur_ind, cur_sear_ind)): processed_label[cur_sear_ind[0], cur_sear_ind[1], cur_sear_ind[2]] = np.argmax(counter) counter = np.zeros((label_size,), dtype=np.uint16) cur_sear_ind = cur_ind counter[sorted_label_voxel_pair[i, 3]] += 1 processed_label[cur_sear_ind[0], cur_sear_ind[1], cur_sear_ind[2]] = np.argmax(counter) return processed_label def collate_fn_BEV(data): data2stack = np.stack([d[0] for d in data]).astype(np.float32) label2stack = np.stack([d[1] for d in data]).astype(np.int) grid_ind_stack = [d[2] for d in data] point_label = [d[3] for d in data] xyz = [d[4] for d in data] return torch.from_numpy(data2stack), torch.from_numpy(label2stack), grid_ind_stack, point_label, xyz # SemKITTI_label_name = {0: 'noise', # 1: 'animal', # 2: 'human.pedestrian.adult', # 3: 'human.pedestrian.child', # 4: 'human.pedestrian.construction_worker', # 5: 'human.pedestrian.personal_mobility', # 6: 'human.pedestrian.police_officer', # 7: 'human.pedestrian.stroller', # 8: 'human.pedestrian.wheelchair', # 9: 'movable_object.barrier', # 10: 'movable_object.debris', # 11: 'movable_object.pushable_pullable', # 12: 'movable_object.trafficcone', # 13: 'static_object.bicycle_rack', # 14: 'vehicle.bicycle', # 15: 'vehicle.bus.bendy', # 16: 'vehicle.bus.rigid', # 17: 'vehicle.car', # 18: 'vehicle.construction', # 19: 'vehicle.emergency.ambulance', # 20: 'vehicle.emergency.police', # 21: 'vehicle.motorcycle', # 22: 'vehicle.trailer', # 23: 'vehicle.truck', # 24: 'flat.driveable_surface', # 25: 'flat.other', # 26: 'flat.sidewalk', # 27: 'flat.terrain', # 28: 'static.manmade', # 29: 'static.other', # 30: 'static.vegetation', # 31: 'vehicle.ego' # } # # SemKITTI_label_name_16 = { # 0: 'noise', # 1: 'barrier', # 2: 'bicycle', # 3: 'bus', # 4: 'car', # 5: 'construction_vehicle', # 6: 'motorcycle', # 7: 'pedestrian', # 8: 'traffic_cone', # 9: 'trailer', # 10: 'truck', # 11: 'driveable_surface', # 12: 'other_flat', # 13: 'sidewalk', # 14: 'terrain', # 15: 'manmade', # 16: 'vegetation', # } # # labels_mapping = { # 1: 0, # 5: 0, # 7: 0, # 8: 0, # 10: 0, # 11: 0, # 13: 0, # 19: 0, # 20: 0, # 0: 0, # 29: 0, # 31: 0, # 9: 1, # 14: 2, # 15: 3, # 16: 3, # 17: 4, # 18: 5, # 21: 6, # 2: 7, # 3: 7, # 4: 7, # 6: 7, # 12: 8, # 22: 9, # 23: 10, # 24: 11, # 25: 12, # 26: 13, # 27: 14, # 28: 15, # 30: 16 # }
9,266
35.920319
114
py
T-Concord3D
T-Concord3D-master/dataloader/__init__.py
# -*- coding:utf-8 -*- # author: Xinge # @file: __init__.py.py from . import dataset_nuscenes
95
15
30
py
T-Concord3D
T-Concord3D-master/dataloader/preprocess.py
# -*- coding:utf-8 -*- # author: Awet H. Gebrehiwot # --------------------------| import torch import torchvision.transforms as transforms from augmentations import RandAugment3D def preprocessing(point_set, cls): pts_transform = transforms.Compose( [] ) pts_transform.transforms.insert(0, RandAugment3D(2, 2)) return torch.from_numpy(pts_transform(point_set)), cls
397
23.875
59
py
darts
darts-master/cnn/test.py
import os import sys import glob import numpy as np import torch import utils import logging import argparse import torch.nn as nn import genotypes import torch.utils import torchvision.datasets as dset import torch.backends.cudnn as cudnn from torch.autograd import Variable from model import NetworkCIFAR as Network parser = argparse.ArgumentParser("cifar") parser.add_argument('--data', type=str, default='../data', help='location of the data corpus') parser.add_argument('--batch_size', type=int, default=96, help='batch size') parser.add_argument('--report_freq', type=float, default=50, help='report frequency') parser.add_argument('--gpu', type=int, default=0, help='gpu device id') parser.add_argument('--init_channels', type=int, default=36, help='num of init channels') parser.add_argument('--layers', type=int, default=20, help='total number of layers') parser.add_argument('--model_path', type=str, default='EXP/model.pt', help='path of pretrained model') parser.add_argument('--auxiliary', action='store_true', default=False, help='use auxiliary tower') parser.add_argument('--cutout', action='store_true', default=False, help='use cutout') parser.add_argument('--cutout_length', type=int, default=16, help='cutout length') parser.add_argument('--drop_path_prob', type=float, default=0.2, help='drop path probability') parser.add_argument('--seed', type=int, default=0, help='random seed') parser.add_argument('--arch', type=str, default='DARTS', help='which architecture to use') args = parser.parse_args() log_format = '%(asctime)s %(message)s' logging.basicConfig(stream=sys.stdout, level=logging.INFO, format=log_format, datefmt='%m/%d %I:%M:%S %p') CIFAR_CLASSES = 10 def main(): if not torch.cuda.is_available(): logging.info('no gpu device available') sys.exit(1) np.random.seed(args.seed) torch.cuda.set_device(args.gpu) cudnn.benchmark = True torch.manual_seed(args.seed) cudnn.enabled=True torch.cuda.manual_seed(args.seed) logging.info('gpu device = %d' % args.gpu) logging.info("args = %s", args) genotype = eval("genotypes.%s" % args.arch) model = Network(args.init_channels, CIFAR_CLASSES, args.layers, args.auxiliary, genotype) model = model.cuda() utils.load(model, args.model_path) logging.info("param size = %fMB", utils.count_parameters_in_MB(model)) criterion = nn.CrossEntropyLoss() criterion = criterion.cuda() _, test_transform = utils._data_transforms_cifar10(args) test_data = dset.CIFAR10(root=args.data, train=False, download=True, transform=test_transform) test_queue = torch.utils.data.DataLoader( test_data, batch_size=args.batch_size, shuffle=False, pin_memory=True, num_workers=2) model.drop_path_prob = args.drop_path_prob test_acc, test_obj = infer(test_queue, model, criterion) logging.info('test_acc %f', test_acc) def infer(test_queue, model, criterion): objs = utils.AvgrageMeter() top1 = utils.AvgrageMeter() top5 = utils.AvgrageMeter() model.eval() for step, (input, target) in enumerate(test_queue): input = Variable(input, volatile=True).cuda() target = Variable(target, volatile=True).cuda(async=True) logits, _ = model(input) loss = criterion(logits, target) prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5)) n = input.size(0) objs.update(loss.data[0], n) top1.update(prec1.data[0], n) top5.update(prec5.data[0], n) if step % args.report_freq == 0: logging.info('test %03d %e %f %f', step, objs.avg, top1.avg, top5.avg) return top1.avg, objs.avg if __name__ == '__main__': main()
3,593
33.228571
102
py
darts
darts-master/cnn/architect.py
import torch import numpy as np import torch.nn as nn from torch.autograd import Variable def _concat(xs): return torch.cat([x.view(-1) for x in xs]) class Architect(object): def __init__(self, model, args): self.network_momentum = args.momentum self.network_weight_decay = args.weight_decay self.model = model self.optimizer = torch.optim.Adam(self.model.arch_parameters(), lr=args.arch_learning_rate, betas=(0.5, 0.999), weight_decay=args.arch_weight_decay) def _compute_unrolled_model(self, input, target, eta, network_optimizer): loss = self.model._loss(input, target) theta = _concat(self.model.parameters()).data try: moment = _concat(network_optimizer.state[v]['momentum_buffer'] for v in self.model.parameters()).mul_(self.network_momentum) except: moment = torch.zeros_like(theta) dtheta = _concat(torch.autograd.grad(loss, self.model.parameters())).data + self.network_weight_decay*theta unrolled_model = self._construct_model_from_theta(theta.sub(eta, moment+dtheta)) return unrolled_model def step(self, input_train, target_train, input_valid, target_valid, eta, network_optimizer, unrolled): self.optimizer.zero_grad() if unrolled: self._backward_step_unrolled(input_train, target_train, input_valid, target_valid, eta, network_optimizer) else: self._backward_step(input_valid, target_valid) self.optimizer.step() def _backward_step(self, input_valid, target_valid): loss = self.model._loss(input_valid, target_valid) loss.backward() def _backward_step_unrolled(self, input_train, target_train, input_valid, target_valid, eta, network_optimizer): unrolled_model = self._compute_unrolled_model(input_train, target_train, eta, network_optimizer) unrolled_loss = unrolled_model._loss(input_valid, target_valid) unrolled_loss.backward() dalpha = [v.grad for v in unrolled_model.arch_parameters()] vector = [v.grad.data for v in unrolled_model.parameters()] implicit_grads = self._hessian_vector_product(vector, input_train, target_train) for g, ig in zip(dalpha, implicit_grads): g.data.sub_(eta, ig.data) for v, g in zip(self.model.arch_parameters(), dalpha): if v.grad is None: v.grad = Variable(g.data) else: v.grad.data.copy_(g.data) def _construct_model_from_theta(self, theta): model_new = self.model.new() model_dict = self.model.state_dict() params, offset = {}, 0 for k, v in self.model.named_parameters(): v_length = np.prod(v.size()) params[k] = theta[offset: offset+v_length].view(v.size()) offset += v_length assert offset == len(theta) model_dict.update(params) model_new.load_state_dict(model_dict) return model_new.cuda() def _hessian_vector_product(self, vector, input, target, r=1e-2): R = r / _concat(vector).norm() for p, v in zip(self.model.parameters(), vector): p.data.add_(R, v) loss = self.model._loss(input, target) grads_p = torch.autograd.grad(loss, self.model.arch_parameters()) for p, v in zip(self.model.parameters(), vector): p.data.sub_(2*R, v) loss = self.model._loss(input, target) grads_n = torch.autograd.grad(loss, self.model.arch_parameters()) for p, v in zip(self.model.parameters(), vector): p.data.add_(R, v) return [(x-y).div_(2*R) for x, y in zip(grads_p, grads_n)]
3,429
35.88172
130
py
darts
darts-master/cnn/train_imagenet.py
import os import sys import numpy as np import time import torch import utils import glob import random import logging import argparse import torch.nn as nn import genotypes import torch.utils import torchvision.datasets as dset import torchvision.transforms as transforms import torch.backends.cudnn as cudnn from torch.autograd import Variable from model import NetworkImageNet as Network parser = argparse.ArgumentParser("imagenet") parser.add_argument('--data', type=str, default='../data/imagenet/', help='location of the data corpus') parser.add_argument('--batch_size', type=int, default=128, help='batch size') parser.add_argument('--learning_rate', type=float, default=0.1, help='init learning rate') parser.add_argument('--momentum', type=float, default=0.9, help='momentum') parser.add_argument('--weight_decay', type=float, default=3e-5, help='weight decay') parser.add_argument('--report_freq', type=float, default=100, help='report frequency') parser.add_argument('--gpu', type=int, default=0, help='gpu device id') parser.add_argument('--epochs', type=int, default=250, help='num of training epochs') parser.add_argument('--init_channels', type=int, default=48, help='num of init channels') parser.add_argument('--layers', type=int, default=14, help='total number of layers') parser.add_argument('--auxiliary', action='store_true', default=False, help='use auxiliary tower') parser.add_argument('--auxiliary_weight', type=float, default=0.4, help='weight for auxiliary loss') parser.add_argument('--drop_path_prob', type=float, default=0, help='drop path probability') parser.add_argument('--save', type=str, default='EXP', help='experiment name') parser.add_argument('--seed', type=int, default=0, help='random seed') parser.add_argument('--arch', type=str, default='DARTS', help='which architecture to use') parser.add_argument('--grad_clip', type=float, default=5., help='gradient clipping') parser.add_argument('--label_smooth', type=float, default=0.1, help='label smoothing') parser.add_argument('--gamma', type=float, default=0.97, help='learning rate decay') parser.add_argument('--decay_period', type=int, default=1, help='epochs between two learning rate decays') parser.add_argument('--parallel', action='store_true', default=False, help='data parallelism') args = parser.parse_args() args.save = 'eval-{}-{}'.format(args.save, time.strftime("%Y%m%d-%H%M%S")) utils.create_exp_dir(args.save, scripts_to_save=glob.glob('*.py')) log_format = '%(asctime)s %(message)s' logging.basicConfig(stream=sys.stdout, level=logging.INFO, format=log_format, datefmt='%m/%d %I:%M:%S %p') fh = logging.FileHandler(os.path.join(args.save, 'log.txt')) fh.setFormatter(logging.Formatter(log_format)) logging.getLogger().addHandler(fh) CLASSES = 1000 class CrossEntropyLabelSmooth(nn.Module): def __init__(self, num_classes, epsilon): super(CrossEntropyLabelSmooth, self).__init__() self.num_classes = num_classes self.epsilon = epsilon self.logsoftmax = nn.LogSoftmax(dim=1) def forward(self, inputs, targets): log_probs = self.logsoftmax(inputs) targets = torch.zeros_like(log_probs).scatter_(1, targets.unsqueeze(1), 1) targets = (1 - self.epsilon) * targets + self.epsilon / self.num_classes loss = (-targets * log_probs).mean(0).sum() return loss def main(): if not torch.cuda.is_available(): logging.info('no gpu device available') sys.exit(1) np.random.seed(args.seed) torch.cuda.set_device(args.gpu) cudnn.benchmark = True torch.manual_seed(args.seed) cudnn.enabled=True torch.cuda.manual_seed(args.seed) logging.info('gpu device = %d' % args.gpu) logging.info("args = %s", args) genotype = eval("genotypes.%s" % args.arch) model = Network(args.init_channels, CLASSES, args.layers, args.auxiliary, genotype) if args.parallel: model = nn.DataParallel(model).cuda() else: model = model.cuda() logging.info("param size = %fMB", utils.count_parameters_in_MB(model)) criterion = nn.CrossEntropyLoss() criterion = criterion.cuda() criterion_smooth = CrossEntropyLabelSmooth(CLASSES, args.label_smooth) criterion_smooth = criterion_smooth.cuda() optimizer = torch.optim.SGD( model.parameters(), args.learning_rate, momentum=args.momentum, weight_decay=args.weight_decay ) traindir = os.path.join(args.data, 'train') validdir = os.path.join(args.data, 'val') normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) train_data = dset.ImageFolder( traindir, transforms.Compose([ transforms.RandomResizedCrop(224), transforms.RandomHorizontalFlip(), transforms.ColorJitter( brightness=0.4, contrast=0.4, saturation=0.4, hue=0.2), transforms.ToTensor(), normalize, ])) valid_data = dset.ImageFolder( validdir, transforms.Compose([ transforms.Resize(256), transforms.CenterCrop(224), transforms.ToTensor(), normalize, ])) train_queue = torch.utils.data.DataLoader( train_data, batch_size=args.batch_size, shuffle=True, pin_memory=True, num_workers=4) valid_queue = torch.utils.data.DataLoader( valid_data, batch_size=args.batch_size, shuffle=False, pin_memory=True, num_workers=4) scheduler = torch.optim.lr_scheduler.StepLR(optimizer, args.decay_period, gamma=args.gamma) best_acc_top1 = 0 for epoch in range(args.epochs): scheduler.step() logging.info('epoch %d lr %e', epoch, scheduler.get_lr()[0]) model.drop_path_prob = args.drop_path_prob * epoch / args.epochs train_acc, train_obj = train(train_queue, model, criterion_smooth, optimizer) logging.info('train_acc %f', train_acc) valid_acc_top1, valid_acc_top5, valid_obj = infer(valid_queue, model, criterion) logging.info('valid_acc_top1 %f', valid_acc_top1) logging.info('valid_acc_top5 %f', valid_acc_top5) is_best = False if valid_acc_top1 > best_acc_top1: best_acc_top1 = valid_acc_top1 is_best = True utils.save_checkpoint({ 'epoch': epoch + 1, 'state_dict': model.state_dict(), 'best_acc_top1': best_acc_top1, 'optimizer' : optimizer.state_dict(), }, is_best, args.save) def train(train_queue, model, criterion, optimizer): objs = utils.AvgrageMeter() top1 = utils.AvgrageMeter() top5 = utils.AvgrageMeter() model.train() for step, (input, target) in enumerate(train_queue): target = target.cuda(async=True) input = input.cuda() input = Variable(input) target = Variable(target) optimizer.zero_grad() logits, logits_aux = model(input) loss = criterion(logits, target) if args.auxiliary: loss_aux = criterion(logits_aux, target) loss += args.auxiliary_weight*loss_aux loss.backward() nn.utils.clip_grad_norm(model.parameters(), args.grad_clip) optimizer.step() prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5)) n = input.size(0) objs.update(loss.data[0], n) top1.update(prec1.data[0], n) top5.update(prec5.data[0], n) if step % args.report_freq == 0: logging.info('train %03d %e %f %f', step, objs.avg, top1.avg, top5.avg) return top1.avg, objs.avg def infer(valid_queue, model, criterion): objs = utils.AvgrageMeter() top1 = utils.AvgrageMeter() top5 = utils.AvgrageMeter() model.eval() for step, (input, target) in enumerate(valid_queue): input = Variable(input, volatile=True).cuda() target = Variable(target, volatile=True).cuda(async=True) logits, _ = model(input) loss = criterion(logits, target) prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5)) n = input.size(0) objs.update(loss.data[0], n) top1.update(prec1.data[0], n) top5.update(prec5.data[0], n) if step % args.report_freq == 0: logging.info('valid %03d %e %f %f', step, objs.avg, top1.avg, top5.avg) return top1.avg, top5.avg, objs.avg if __name__ == '__main__': main()
7,992
33.601732
106
py
darts
darts-master/cnn/utils.py
import os import numpy as np import torch import shutil import torchvision.transforms as transforms from torch.autograd import Variable class AvgrageMeter(object): def __init__(self): self.reset() def reset(self): self.avg = 0 self.sum = 0 self.cnt = 0 def update(self, val, n=1): self.sum += val * n self.cnt += n self.avg = self.sum / self.cnt def accuracy(output, target, topk=(1,)): maxk = max(topk) batch_size = target.size(0) _, pred = output.topk(maxk, 1, True, True) pred = pred.t() correct = pred.eq(target.view(1, -1).expand_as(pred)) res = [] for k in topk: correct_k = correct[:k].view(-1).float().sum(0) res.append(correct_k.mul_(100.0/batch_size)) return res class Cutout(object): def __init__(self, length): self.length = length def __call__(self, img): h, w = img.size(1), img.size(2) mask = np.ones((h, w), np.float32) y = np.random.randint(h) x = np.random.randint(w) y1 = np.clip(y - self.length // 2, 0, h) y2 = np.clip(y + self.length // 2, 0, h) x1 = np.clip(x - self.length // 2, 0, w) x2 = np.clip(x + self.length // 2, 0, w) mask[y1: y2, x1: x2] = 0. mask = torch.from_numpy(mask) mask = mask.expand_as(img) img *= mask return img def _data_transforms_cifar10(args): CIFAR_MEAN = [0.49139968, 0.48215827, 0.44653124] CIFAR_STD = [0.24703233, 0.24348505, 0.26158768] train_transform = transforms.Compose([ transforms.RandomCrop(32, padding=4), transforms.RandomHorizontalFlip(), transforms.ToTensor(), transforms.Normalize(CIFAR_MEAN, CIFAR_STD), ]) if args.cutout: train_transform.transforms.append(Cutout(args.cutout_length)) valid_transform = transforms.Compose([ transforms.ToTensor(), transforms.Normalize(CIFAR_MEAN, CIFAR_STD), ]) return train_transform, valid_transform def count_parameters_in_MB(model): return np.sum(np.prod(v.size()) for name, v in model.named_parameters() if "auxiliary" not in name)/1e6 def save_checkpoint(state, is_best, save): filename = os.path.join(save, 'checkpoint.pth.tar') torch.save(state, filename) if is_best: best_filename = os.path.join(save, 'model_best.pth.tar') shutil.copyfile(filename, best_filename) def save(model, model_path): torch.save(model.state_dict(), model_path) def load(model, model_path): model.load_state_dict(torch.load(model_path)) def drop_path(x, drop_prob): if drop_prob > 0.: keep_prob = 1.-drop_prob mask = Variable(torch.cuda.FloatTensor(x.size(0), 1, 1, 1).bernoulli_(keep_prob)) x.div_(keep_prob) x.mul_(mask) return x def create_exp_dir(path, scripts_to_save=None): if not os.path.exists(path): os.mkdir(path) print('Experiment dir : {}'.format(path)) if scripts_to_save is not None: os.mkdir(os.path.join(path, 'scripts')) for script in scripts_to_save: dst_file = os.path.join(path, 'scripts', os.path.basename(script)) shutil.copyfile(script, dst_file)
3,080
24.254098
105
py
darts
darts-master/cnn/model.py
import torch import torch.nn as nn from operations import * from torch.autograd import Variable from utils import drop_path class Cell(nn.Module): def __init__(self, genotype, C_prev_prev, C_prev, C, reduction, reduction_prev): super(Cell, self).__init__() print(C_prev_prev, C_prev, C) if reduction_prev: self.preprocess0 = FactorizedReduce(C_prev_prev, C) else: self.preprocess0 = ReLUConvBN(C_prev_prev, C, 1, 1, 0) self.preprocess1 = ReLUConvBN(C_prev, C, 1, 1, 0) if reduction: op_names, indices = zip(*genotype.reduce) concat = genotype.reduce_concat else: op_names, indices = zip(*genotype.normal) concat = genotype.normal_concat self._compile(C, op_names, indices, concat, reduction) def _compile(self, C, op_names, indices, concat, reduction): assert len(op_names) == len(indices) self._steps = len(op_names) // 2 self._concat = concat self.multiplier = len(concat) self._ops = nn.ModuleList() for name, index in zip(op_names, indices): stride = 2 if reduction and index < 2 else 1 op = OPS[name](C, stride, True) self._ops += [op] self._indices = indices def forward(self, s0, s1, drop_prob): s0 = self.preprocess0(s0) s1 = self.preprocess1(s1) states = [s0, s1] for i in range(self._steps): h1 = states[self._indices[2*i]] h2 = states[self._indices[2*i+1]] op1 = self._ops[2*i] op2 = self._ops[2*i+1] h1 = op1(h1) h2 = op2(h2) if self.training and drop_prob > 0.: if not isinstance(op1, Identity): h1 = drop_path(h1, drop_prob) if not isinstance(op2, Identity): h2 = drop_path(h2, drop_prob) s = h1 + h2 states += [s] return torch.cat([states[i] for i in self._concat], dim=1) class AuxiliaryHeadCIFAR(nn.Module): def __init__(self, C, num_classes): """assuming input size 8x8""" super(AuxiliaryHeadCIFAR, self).__init__() self.features = nn.Sequential( nn.ReLU(inplace=True), nn.AvgPool2d(5, stride=3, padding=0, count_include_pad=False), # image size = 2 x 2 nn.Conv2d(C, 128, 1, bias=False), nn.BatchNorm2d(128), nn.ReLU(inplace=True), nn.Conv2d(128, 768, 2, bias=False), nn.BatchNorm2d(768), nn.ReLU(inplace=True) ) self.classifier = nn.Linear(768, num_classes) def forward(self, x): x = self.features(x) x = self.classifier(x.view(x.size(0),-1)) return x class AuxiliaryHeadImageNet(nn.Module): def __init__(self, C, num_classes): """assuming input size 14x14""" super(AuxiliaryHeadImageNet, self).__init__() self.features = nn.Sequential( nn.ReLU(inplace=True), nn.AvgPool2d(5, stride=2, padding=0, count_include_pad=False), nn.Conv2d(C, 128, 1, bias=False), nn.BatchNorm2d(128), nn.ReLU(inplace=True), nn.Conv2d(128, 768, 2, bias=False), # NOTE: This batchnorm was omitted in my earlier implementation due to a typo. # Commenting it out for consistency with the experiments in the paper. # nn.BatchNorm2d(768), nn.ReLU(inplace=True) ) self.classifier = nn.Linear(768, num_classes) def forward(self, x): x = self.features(x) x = self.classifier(x.view(x.size(0),-1)) return x class NetworkCIFAR(nn.Module): def __init__(self, C, num_classes, layers, auxiliary, genotype): super(NetworkCIFAR, self).__init__() self._layers = layers self._auxiliary = auxiliary stem_multiplier = 3 C_curr = stem_multiplier*C self.stem = nn.Sequential( nn.Conv2d(3, C_curr, 3, padding=1, bias=False), nn.BatchNorm2d(C_curr) ) C_prev_prev, C_prev, C_curr = C_curr, C_curr, C self.cells = nn.ModuleList() reduction_prev = False for i in range(layers): if i in [layers//3, 2*layers//3]: C_curr *= 2 reduction = True else: reduction = False cell = Cell(genotype, C_prev_prev, C_prev, C_curr, reduction, reduction_prev) reduction_prev = reduction self.cells += [cell] C_prev_prev, C_prev = C_prev, cell.multiplier*C_curr if i == 2*layers//3: C_to_auxiliary = C_prev if auxiliary: self.auxiliary_head = AuxiliaryHeadCIFAR(C_to_auxiliary, num_classes) self.global_pooling = nn.AdaptiveAvgPool2d(1) self.classifier = nn.Linear(C_prev, num_classes) def forward(self, input): logits_aux = None s0 = s1 = self.stem(input) for i, cell in enumerate(self.cells): s0, s1 = s1, cell(s0, s1, self.drop_path_prob) if i == 2*self._layers//3: if self._auxiliary and self.training: logits_aux = self.auxiliary_head(s1) out = self.global_pooling(s1) logits = self.classifier(out.view(out.size(0),-1)) return logits, logits_aux class NetworkImageNet(nn.Module): def __init__(self, C, num_classes, layers, auxiliary, genotype): super(NetworkImageNet, self).__init__() self._layers = layers self._auxiliary = auxiliary self.stem0 = nn.Sequential( nn.Conv2d(3, C // 2, kernel_size=3, stride=2, padding=1, bias=False), nn.BatchNorm2d(C // 2), nn.ReLU(inplace=True), nn.Conv2d(C // 2, C, 3, stride=2, padding=1, bias=False), nn.BatchNorm2d(C), ) self.stem1 = nn.Sequential( nn.ReLU(inplace=True), nn.Conv2d(C, C, 3, stride=2, padding=1, bias=False), nn.BatchNorm2d(C), ) C_prev_prev, C_prev, C_curr = C, C, C self.cells = nn.ModuleList() reduction_prev = True for i in range(layers): if i in [layers // 3, 2 * layers // 3]: C_curr *= 2 reduction = True else: reduction = False cell = Cell(genotype, C_prev_prev, C_prev, C_curr, reduction, reduction_prev) reduction_prev = reduction self.cells += [cell] C_prev_prev, C_prev = C_prev, cell.multiplier * C_curr if i == 2 * layers // 3: C_to_auxiliary = C_prev if auxiliary: self.auxiliary_head = AuxiliaryHeadImageNet(C_to_auxiliary, num_classes) self.global_pooling = nn.AvgPool2d(7) self.classifier = nn.Linear(C_prev, num_classes) def forward(self, input): logits_aux = None s0 = self.stem0(input) s1 = self.stem1(s0) for i, cell in enumerate(self.cells): s0, s1 = s1, cell(s0, s1, self.drop_path_prob) if i == 2 * self._layers // 3: if self._auxiliary and self.training: logits_aux = self.auxiliary_head(s1) out = self.global_pooling(s1) logits = self.classifier(out.view(out.size(0), -1)) return logits, logits_aux
6,640
29.888372
89
py
darts
darts-master/cnn/model_search.py
import torch import torch.nn as nn import torch.nn.functional as F from operations import * from torch.autograd import Variable from genotypes import PRIMITIVES from genotypes import Genotype class MixedOp(nn.Module): def __init__(self, C, stride): super(MixedOp, self).__init__() self._ops = nn.ModuleList() for primitive in PRIMITIVES: op = OPS[primitive](C, stride, False) if 'pool' in primitive: op = nn.Sequential(op, nn.BatchNorm2d(C, affine=False)) self._ops.append(op) def forward(self, x, weights): return sum(w * op(x) for w, op in zip(weights, self._ops)) class Cell(nn.Module): def __init__(self, steps, multiplier, C_prev_prev, C_prev, C, reduction, reduction_prev): super(Cell, self).__init__() self.reduction = reduction if reduction_prev: self.preprocess0 = FactorizedReduce(C_prev_prev, C, affine=False) else: self.preprocess0 = ReLUConvBN(C_prev_prev, C, 1, 1, 0, affine=False) self.preprocess1 = ReLUConvBN(C_prev, C, 1, 1, 0, affine=False) self._steps = steps self._multiplier = multiplier self._ops = nn.ModuleList() self._bns = nn.ModuleList() for i in range(self._steps): for j in range(2+i): stride = 2 if reduction and j < 2 else 1 op = MixedOp(C, stride) self._ops.append(op) def forward(self, s0, s1, weights): s0 = self.preprocess0(s0) s1 = self.preprocess1(s1) states = [s0, s1] offset = 0 for i in range(self._steps): s = sum(self._ops[offset+j](h, weights[offset+j]) for j, h in enumerate(states)) offset += len(states) states.append(s) return torch.cat(states[-self._multiplier:], dim=1) class Network(nn.Module): def __init__(self, C, num_classes, layers, criterion, steps=4, multiplier=4, stem_multiplier=3): super(Network, self).__init__() self._C = C self._num_classes = num_classes self._layers = layers self._criterion = criterion self._steps = steps self._multiplier = multiplier C_curr = stem_multiplier*C self.stem = nn.Sequential( nn.Conv2d(3, C_curr, 3, padding=1, bias=False), nn.BatchNorm2d(C_curr) ) C_prev_prev, C_prev, C_curr = C_curr, C_curr, C self.cells = nn.ModuleList() reduction_prev = False for i in range(layers): if i in [layers//3, 2*layers//3]: C_curr *= 2 reduction = True else: reduction = False cell = Cell(steps, multiplier, C_prev_prev, C_prev, C_curr, reduction, reduction_prev) reduction_prev = reduction self.cells += [cell] C_prev_prev, C_prev = C_prev, multiplier*C_curr self.global_pooling = nn.AdaptiveAvgPool2d(1) self.classifier = nn.Linear(C_prev, num_classes) self._initialize_alphas() def new(self): model_new = Network(self._C, self._num_classes, self._layers, self._criterion).cuda() for x, y in zip(model_new.arch_parameters(), self.arch_parameters()): x.data.copy_(y.data) return model_new def forward(self, input): s0 = s1 = self.stem(input) for i, cell in enumerate(self.cells): if cell.reduction: weights = F.softmax(self.alphas_reduce, dim=-1) else: weights = F.softmax(self.alphas_normal, dim=-1) s0, s1 = s1, cell(s0, s1, weights) out = self.global_pooling(s1) logits = self.classifier(out.view(out.size(0),-1)) return logits def _loss(self, input, target): logits = self(input) return self._criterion(logits, target) def _initialize_alphas(self): k = sum(1 for i in range(self._steps) for n in range(2+i)) num_ops = len(PRIMITIVES) self.alphas_normal = Variable(1e-3*torch.randn(k, num_ops).cuda(), requires_grad=True) self.alphas_reduce = Variable(1e-3*torch.randn(k, num_ops).cuda(), requires_grad=True) self._arch_parameters = [ self.alphas_normal, self.alphas_reduce, ] def arch_parameters(self): return self._arch_parameters def genotype(self): def _parse(weights): gene = [] n = 2 start = 0 for i in range(self._steps): end = start + n W = weights[start:end].copy() edges = sorted(range(i + 2), key=lambda x: -max(W[x][k] for k in range(len(W[x])) if k != PRIMITIVES.index('none')))[:2] for j in edges: k_best = None for k in range(len(W[j])): if k != PRIMITIVES.index('none'): if k_best is None or W[j][k] > W[j][k_best]: k_best = k gene.append((PRIMITIVES[k_best], j)) start = end n += 1 return gene gene_normal = _parse(F.softmax(self.alphas_normal, dim=-1).data.cpu().numpy()) gene_reduce = _parse(F.softmax(self.alphas_reduce, dim=-1).data.cpu().numpy()) concat = range(2+self._steps-self._multiplier, self._steps+2) genotype = Genotype( normal=gene_normal, normal_concat=concat, reduce=gene_reduce, reduce_concat=concat ) return genotype
5,009
29.54878
128
py
darts
darts-master/cnn/train_search.py
import os import sys import time import glob import numpy as np import torch import utils import logging import argparse import torch.nn as nn import torch.utils import torch.nn.functional as F import torchvision.datasets as dset import torch.backends.cudnn as cudnn from torch.autograd import Variable from model_search import Network from architect import Architect parser = argparse.ArgumentParser("cifar") parser.add_argument('--data', type=str, default='../data', help='location of the data corpus') parser.add_argument('--batch_size', type=int, default=64, help='batch size') parser.add_argument('--learning_rate', type=float, default=0.025, help='init learning rate') parser.add_argument('--learning_rate_min', type=float, default=0.001, help='min learning rate') parser.add_argument('--momentum', type=float, default=0.9, help='momentum') parser.add_argument('--weight_decay', type=float, default=3e-4, help='weight decay') parser.add_argument('--report_freq', type=float, default=50, help='report frequency') parser.add_argument('--gpu', type=int, default=0, help='gpu device id') parser.add_argument('--epochs', type=int, default=50, help='num of training epochs') parser.add_argument('--init_channels', type=int, default=16, help='num of init channels') parser.add_argument('--layers', type=int, default=8, help='total number of layers') parser.add_argument('--model_path', type=str, default='saved_models', help='path to save the model') parser.add_argument('--cutout', action='store_true', default=False, help='use cutout') parser.add_argument('--cutout_length', type=int, default=16, help='cutout length') parser.add_argument('--drop_path_prob', type=float, default=0.3, help='drop path probability') parser.add_argument('--save', type=str, default='EXP', help='experiment name') parser.add_argument('--seed', type=int, default=2, help='random seed') parser.add_argument('--grad_clip', type=float, default=5, help='gradient clipping') parser.add_argument('--train_portion', type=float, default=0.5, help='portion of training data') parser.add_argument('--unrolled', action='store_true', default=False, help='use one-step unrolled validation loss') parser.add_argument('--arch_learning_rate', type=float, default=3e-4, help='learning rate for arch encoding') parser.add_argument('--arch_weight_decay', type=float, default=1e-3, help='weight decay for arch encoding') args = parser.parse_args() args.save = 'search-{}-{}'.format(args.save, time.strftime("%Y%m%d-%H%M%S")) utils.create_exp_dir(args.save, scripts_to_save=glob.glob('*.py')) log_format = '%(asctime)s %(message)s' logging.basicConfig(stream=sys.stdout, level=logging.INFO, format=log_format, datefmt='%m/%d %I:%M:%S %p') fh = logging.FileHandler(os.path.join(args.save, 'log.txt')) fh.setFormatter(logging.Formatter(log_format)) logging.getLogger().addHandler(fh) CIFAR_CLASSES = 10 def main(): if not torch.cuda.is_available(): logging.info('no gpu device available') sys.exit(1) np.random.seed(args.seed) torch.cuda.set_device(args.gpu) cudnn.benchmark = True torch.manual_seed(args.seed) cudnn.enabled=True torch.cuda.manual_seed(args.seed) logging.info('gpu device = %d' % args.gpu) logging.info("args = %s", args) criterion = nn.CrossEntropyLoss() criterion = criterion.cuda() model = Network(args.init_channels, CIFAR_CLASSES, args.layers, criterion) model = model.cuda() logging.info("param size = %fMB", utils.count_parameters_in_MB(model)) optimizer = torch.optim.SGD( model.parameters(), args.learning_rate, momentum=args.momentum, weight_decay=args.weight_decay) train_transform, valid_transform = utils._data_transforms_cifar10(args) train_data = dset.CIFAR10(root=args.data, train=True, download=True, transform=train_transform) num_train = len(train_data) indices = list(range(num_train)) split = int(np.floor(args.train_portion * num_train)) train_queue = torch.utils.data.DataLoader( train_data, batch_size=args.batch_size, sampler=torch.utils.data.sampler.SubsetRandomSampler(indices[:split]), pin_memory=True, num_workers=2) valid_queue = torch.utils.data.DataLoader( train_data, batch_size=args.batch_size, sampler=torch.utils.data.sampler.SubsetRandomSampler(indices[split:num_train]), pin_memory=True, num_workers=2) scheduler = torch.optim.lr_scheduler.CosineAnnealingLR( optimizer, float(args.epochs), eta_min=args.learning_rate_min) architect = Architect(model, args) for epoch in range(args.epochs): scheduler.step() lr = scheduler.get_lr()[0] logging.info('epoch %d lr %e', epoch, lr) genotype = model.genotype() logging.info('genotype = %s', genotype) print(F.softmax(model.alphas_normal, dim=-1)) print(F.softmax(model.alphas_reduce, dim=-1)) # training train_acc, train_obj = train(train_queue, valid_queue, model, architect, criterion, optimizer, lr) logging.info('train_acc %f', train_acc) # validation valid_acc, valid_obj = infer(valid_queue, model, criterion) logging.info('valid_acc %f', valid_acc) utils.save(model, os.path.join(args.save, 'weights.pt')) def train(train_queue, valid_queue, model, architect, criterion, optimizer, lr): objs = utils.AvgrageMeter() top1 = utils.AvgrageMeter() top5 = utils.AvgrageMeter() for step, (input, target) in enumerate(train_queue): model.train() n = input.size(0) input = Variable(input, requires_grad=False).cuda() target = Variable(target, requires_grad=False).cuda(async=True) # get a random minibatch from the search queue with replacement input_search, target_search = next(iter(valid_queue)) input_search = Variable(input_search, requires_grad=False).cuda() target_search = Variable(target_search, requires_grad=False).cuda(async=True) architect.step(input, target, input_search, target_search, lr, optimizer, unrolled=args.unrolled) optimizer.zero_grad() logits = model(input) loss = criterion(logits, target) loss.backward() nn.utils.clip_grad_norm(model.parameters(), args.grad_clip) optimizer.step() prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5)) objs.update(loss.data[0], n) top1.update(prec1.data[0], n) top5.update(prec5.data[0], n) if step % args.report_freq == 0: logging.info('train %03d %e %f %f', step, objs.avg, top1.avg, top5.avg) return top1.avg, objs.avg def infer(valid_queue, model, criterion): objs = utils.AvgrageMeter() top1 = utils.AvgrageMeter() top5 = utils.AvgrageMeter() model.eval() for step, (input, target) in enumerate(valid_queue): input = Variable(input, volatile=True).cuda() target = Variable(target, volatile=True).cuda(async=True) logits = model(input) loss = criterion(logits, target) prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5)) n = input.size(0) objs.update(loss.data[0], n) top1.update(prec1.data[0], n) top5.update(prec5.data[0], n) if step % args.report_freq == 0: logging.info('valid %03d %e %f %f', step, objs.avg, top1.avg, top5.avg) return top1.avg, objs.avg if __name__ == '__main__': main()
7,212
35.80102
115
py
darts
darts-master/cnn/test_imagenet.py
import os import sys import numpy as np import torch import utils import glob import random import logging import argparse import torch.nn as nn import genotypes import torch.utils import torchvision.datasets as dset import torchvision.transforms as transforms import torch.backends.cudnn as cudnn from torch.autograd import Variable from model import NetworkImageNet as Network parser = argparse.ArgumentParser("imagenet") parser.add_argument('--data', type=str, default='../data/imagenet/', help='location of the data corpus') parser.add_argument('--batch_size', type=int, default=128, help='batch size') parser.add_argument('--report_freq', type=float, default=100, help='report frequency') parser.add_argument('--gpu', type=int, default=0, help='gpu device id') parser.add_argument('--init_channels', type=int, default=48, help='num of init channels') parser.add_argument('--layers', type=int, default=14, help='total number of layers') parser.add_argument('--model_path', type=str, default='EXP/model.pt', help='path of pretrained model') parser.add_argument('--auxiliary', action='store_true', default=False, help='use auxiliary tower') parser.add_argument('--drop_path_prob', type=float, default=0, help='drop path probability') parser.add_argument('--seed', type=int, default=0, help='random seed') parser.add_argument('--arch', type=str, default='DARTS', help='which architecture to use') args = parser.parse_args() log_format = '%(asctime)s %(message)s' logging.basicConfig(stream=sys.stdout, level=logging.INFO, format=log_format, datefmt='%m/%d %I:%M:%S %p') CLASSES = 1000 def main(): if not torch.cuda.is_available(): logging.info('no gpu device available') sys.exit(1) np.random.seed(args.seed) torch.cuda.set_device(args.gpu) cudnn.benchmark = True torch.manual_seed(args.seed) cudnn.enabled=True torch.cuda.manual_seed(args.seed) logging.info('gpu device = %d' % args.gpu) logging.info("args = %s", args) genotype = eval("genotypes.%s" % args.arch) model = Network(args.init_channels, CLASSES, args.layers, args.auxiliary, genotype) model = model.cuda() model.load_state_dict(torch.load(args.model_path)['state_dict']) logging.info("param size = %fMB", utils.count_parameters_in_MB(model)) criterion = nn.CrossEntropyLoss() criterion = criterion.cuda() validdir = os.path.join(args.data, 'val') normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) valid_data = dset.ImageFolder( validdir, transforms.Compose([ transforms.Resize(256), transforms.CenterCrop(224), transforms.ToTensor(), normalize, ])) valid_queue = torch.utils.data.DataLoader( valid_data, batch_size=args.batch_size, shuffle=False, pin_memory=True, num_workers=4) model.drop_path_prob = args.drop_path_prob valid_acc_top1, valid_acc_top5, valid_obj = infer(valid_queue, model, criterion) logging.info('valid_acc_top1 %f', valid_acc_top1) logging.info('valid_acc_top5 %f', valid_acc_top5) def infer(valid_queue, model, criterion): objs = utils.AvgrageMeter() top1 = utils.AvgrageMeter() top5 = utils.AvgrageMeter() model.eval() for step, (input, target) in enumerate(valid_queue): input = Variable(input, volatile=True).cuda() target = Variable(target, volatile=True).cuda(async=True) logits, _ = model(input) loss = criterion(logits, target) prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5)) n = input.size(0) objs.update(loss.data[0], n) top1.update(prec1.data[0], n) top5.update(prec5.data[0], n) if step % args.report_freq == 0: logging.info('valid %03d %e %f %f', step, objs.avg, top1.avg, top5.avg) return top1.avg, top5.avg, objs.avg if __name__ == '__main__': main()
3,785
32.504425
104
py
darts
darts-master/cnn/train.py
import os import sys import time import glob import numpy as np import torch import utils import logging import argparse import torch.nn as nn import genotypes import torch.utils import torchvision.datasets as dset import torch.backends.cudnn as cudnn from torch.autograd import Variable from model import NetworkCIFAR as Network parser = argparse.ArgumentParser("cifar") parser.add_argument('--data', type=str, default='../data', help='location of the data corpus') parser.add_argument('--batch_size', type=int, default=96, help='batch size') parser.add_argument('--learning_rate', type=float, default=0.025, help='init learning rate') parser.add_argument('--momentum', type=float, default=0.9, help='momentum') parser.add_argument('--weight_decay', type=float, default=3e-4, help='weight decay') parser.add_argument('--report_freq', type=float, default=50, help='report frequency') parser.add_argument('--gpu', type=int, default=0, help='gpu device id') parser.add_argument('--epochs', type=int, default=600, help='num of training epochs') parser.add_argument('--init_channels', type=int, default=36, help='num of init channels') parser.add_argument('--layers', type=int, default=20, help='total number of layers') parser.add_argument('--model_path', type=str, default='saved_models', help='path to save the model') parser.add_argument('--auxiliary', action='store_true', default=False, help='use auxiliary tower') parser.add_argument('--auxiliary_weight', type=float, default=0.4, help='weight for auxiliary loss') parser.add_argument('--cutout', action='store_true', default=False, help='use cutout') parser.add_argument('--cutout_length', type=int, default=16, help='cutout length') parser.add_argument('--drop_path_prob', type=float, default=0.2, help='drop path probability') parser.add_argument('--save', type=str, default='EXP', help='experiment name') parser.add_argument('--seed', type=int, default=0, help='random seed') parser.add_argument('--arch', type=str, default='DARTS', help='which architecture to use') parser.add_argument('--grad_clip', type=float, default=5, help='gradient clipping') args = parser.parse_args() args.save = 'eval-{}-{}'.format(args.save, time.strftime("%Y%m%d-%H%M%S")) utils.create_exp_dir(args.save, scripts_to_save=glob.glob('*.py')) log_format = '%(asctime)s %(message)s' logging.basicConfig(stream=sys.stdout, level=logging.INFO, format=log_format, datefmt='%m/%d %I:%M:%S %p') fh = logging.FileHandler(os.path.join(args.save, 'log.txt')) fh.setFormatter(logging.Formatter(log_format)) logging.getLogger().addHandler(fh) CIFAR_CLASSES = 10 def main(): if not torch.cuda.is_available(): logging.info('no gpu device available') sys.exit(1) np.random.seed(args.seed) torch.cuda.set_device(args.gpu) cudnn.benchmark = True torch.manual_seed(args.seed) cudnn.enabled=True torch.cuda.manual_seed(args.seed) logging.info('gpu device = %d' % args.gpu) logging.info("args = %s", args) genotype = eval("genotypes.%s" % args.arch) model = Network(args.init_channels, CIFAR_CLASSES, args.layers, args.auxiliary, genotype) model = model.cuda() logging.info("param size = %fMB", utils.count_parameters_in_MB(model)) criterion = nn.CrossEntropyLoss() criterion = criterion.cuda() optimizer = torch.optim.SGD( model.parameters(), args.learning_rate, momentum=args.momentum, weight_decay=args.weight_decay ) train_transform, valid_transform = utils._data_transforms_cifar10(args) train_data = dset.CIFAR10(root=args.data, train=True, download=True, transform=train_transform) valid_data = dset.CIFAR10(root=args.data, train=False, download=True, transform=valid_transform) train_queue = torch.utils.data.DataLoader( train_data, batch_size=args.batch_size, shuffle=True, pin_memory=True, num_workers=2) valid_queue = torch.utils.data.DataLoader( valid_data, batch_size=args.batch_size, shuffle=False, pin_memory=True, num_workers=2) scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, float(args.epochs)) for epoch in range(args.epochs): scheduler.step() logging.info('epoch %d lr %e', epoch, scheduler.get_lr()[0]) model.drop_path_prob = args.drop_path_prob * epoch / args.epochs train_acc, train_obj = train(train_queue, model, criterion, optimizer) logging.info('train_acc %f', train_acc) valid_acc, valid_obj = infer(valid_queue, model, criterion) logging.info('valid_acc %f', valid_acc) utils.save(model, os.path.join(args.save, 'weights.pt')) def train(train_queue, model, criterion, optimizer): objs = utils.AvgrageMeter() top1 = utils.AvgrageMeter() top5 = utils.AvgrageMeter() model.train() for step, (input, target) in enumerate(train_queue): input = Variable(input).cuda() target = Variable(target).cuda(async=True) optimizer.zero_grad() logits, logits_aux = model(input) loss = criterion(logits, target) if args.auxiliary: loss_aux = criterion(logits_aux, target) loss += args.auxiliary_weight*loss_aux loss.backward() nn.utils.clip_grad_norm(model.parameters(), args.grad_clip) optimizer.step() prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5)) n = input.size(0) objs.update(loss.data[0], n) top1.update(prec1.data[0], n) top5.update(prec5.data[0], n) if step % args.report_freq == 0: logging.info('train %03d %e %f %f', step, objs.avg, top1.avg, top5.avg) return top1.avg, objs.avg def infer(valid_queue, model, criterion): objs = utils.AvgrageMeter() top1 = utils.AvgrageMeter() top5 = utils.AvgrageMeter() model.eval() for step, (input, target) in enumerate(valid_queue): input = Variable(input, volatile=True).cuda() target = Variable(target, volatile=True).cuda(async=True) logits, _ = model(input) loss = criterion(logits, target) prec1, prec5 = utils.accuracy(logits, target, topk=(1, 5)) n = input.size(0) objs.update(loss.data[0], n) top1.update(prec1.data[0], n) top5.update(prec5.data[0], n) if step % args.report_freq == 0: logging.info('valid %03d %e %f %f', step, objs.avg, top1.avg, top5.avg) return top1.avg, objs.avg if __name__ == '__main__': main()
6,251
35.561404
100
py
darts
darts-master/cnn/visualize.py
import sys import genotypes from graphviz import Digraph def plot(genotype, filename): g = Digraph( format='pdf', edge_attr=dict(fontsize='20', fontname="times"), node_attr=dict(style='filled', shape='rect', align='center', fontsize='20', height='0.5', width='0.5', penwidth='2', fontname="times"), engine='dot') g.body.extend(['rankdir=LR']) g.node("c_{k-2}", fillcolor='darkseagreen2') g.node("c_{k-1}", fillcolor='darkseagreen2') assert len(genotype) % 2 == 0 steps = len(genotype) // 2 for i in range(steps): g.node(str(i), fillcolor='lightblue') for i in range(steps): for k in [2*i, 2*i + 1]: op, j = genotype[k] if j == 0: u = "c_{k-2}" elif j == 1: u = "c_{k-1}" else: u = str(j-2) v = str(i) g.edge(u, v, label=op, fillcolor="gray") g.node("c_{k}", fillcolor='palegoldenrod') for i in range(steps): g.edge(str(i), "c_{k}", fillcolor="gray") g.render(filename, view=True) if __name__ == '__main__': if len(sys.argv) != 2: print("usage:\n python {} ARCH_NAME".format(sys.argv[0])) sys.exit(1) genotype_name = sys.argv[1] try: genotype = eval('genotypes.{}'.format(genotype_name)) except AttributeError: print("{} is not specified in genotypes.py".format(genotype_name)) sys.exit(1) plot(genotype.normal, "normal") plot(genotype.reduce, "reduction")
1,419
24.357143
141
py
darts
darts-master/cnn/genotypes.py
from collections import namedtuple Genotype = namedtuple('Genotype', 'normal normal_concat reduce reduce_concat') PRIMITIVES = [ 'none', 'max_pool_3x3', 'avg_pool_3x3', 'skip_connect', 'sep_conv_3x3', 'sep_conv_5x5', 'dil_conv_3x3', 'dil_conv_5x5' ] NASNet = Genotype( normal = [ ('sep_conv_5x5', 1), ('sep_conv_3x3', 0), ('sep_conv_5x5', 0), ('sep_conv_3x3', 0), ('avg_pool_3x3', 1), ('skip_connect', 0), ('avg_pool_3x3', 0), ('avg_pool_3x3', 0), ('sep_conv_3x3', 1), ('skip_connect', 1), ], normal_concat = [2, 3, 4, 5, 6], reduce = [ ('sep_conv_5x5', 1), ('sep_conv_7x7', 0), ('max_pool_3x3', 1), ('sep_conv_7x7', 0), ('avg_pool_3x3', 1), ('sep_conv_5x5', 0), ('skip_connect', 3), ('avg_pool_3x3', 2), ('sep_conv_3x3', 2), ('max_pool_3x3', 1), ], reduce_concat = [4, 5, 6], ) AmoebaNet = Genotype( normal = [ ('avg_pool_3x3', 0), ('max_pool_3x3', 1), ('sep_conv_3x3', 0), ('sep_conv_5x5', 2), ('sep_conv_3x3', 0), ('avg_pool_3x3', 3), ('sep_conv_3x3', 1), ('skip_connect', 1), ('skip_connect', 0), ('avg_pool_3x3', 1), ], normal_concat = [4, 5, 6], reduce = [ ('avg_pool_3x3', 0), ('sep_conv_3x3', 1), ('max_pool_3x3', 0), ('sep_conv_7x7', 2), ('sep_conv_7x7', 0), ('avg_pool_3x3', 1), ('max_pool_3x3', 0), ('max_pool_3x3', 1), ('conv_7x1_1x7', 0), ('sep_conv_3x3', 5), ], reduce_concat = [3, 4, 6] ) DARTS_V1 = Genotype(normal=[('sep_conv_3x3', 1), ('sep_conv_3x3', 0), ('skip_connect', 0), ('sep_conv_3x3', 1), ('skip_connect', 0), ('sep_conv_3x3', 1), ('sep_conv_3x3', 0), ('skip_connect', 2)], normal_concat=[2, 3, 4, 5], reduce=[('max_pool_3x3', 0), ('max_pool_3x3', 1), ('skip_connect', 2), ('max_pool_3x3', 0), ('max_pool_3x3', 0), ('skip_connect', 2), ('skip_connect', 2), ('avg_pool_3x3', 0)], reduce_concat=[2, 3, 4, 5]) DARTS_V2 = Genotype(normal=[('sep_conv_3x3', 0), ('sep_conv_3x3', 1), ('sep_conv_3x3', 0), ('sep_conv_3x3', 1), ('sep_conv_3x3', 1), ('skip_connect', 0), ('skip_connect', 0), ('dil_conv_3x3', 2)], normal_concat=[2, 3, 4, 5], reduce=[('max_pool_3x3', 0), ('max_pool_3x3', 1), ('skip_connect', 2), ('max_pool_3x3', 1), ('max_pool_3x3', 0), ('skip_connect', 2), ('skip_connect', 2), ('max_pool_3x3', 1)], reduce_concat=[2, 3, 4, 5]) DARTS = DARTS_V2
2,410
29.518987
429
py
darts
darts-master/cnn/operations.py
import torch import torch.nn as nn OPS = { 'none' : lambda C, stride, affine: Zero(stride), 'avg_pool_3x3' : lambda C, stride, affine: nn.AvgPool2d(3, stride=stride, padding=1, count_include_pad=False), 'max_pool_3x3' : lambda C, stride, affine: nn.MaxPool2d(3, stride=stride, padding=1), 'skip_connect' : lambda C, stride, affine: Identity() if stride == 1 else FactorizedReduce(C, C, affine=affine), 'sep_conv_3x3' : lambda C, stride, affine: SepConv(C, C, 3, stride, 1, affine=affine), 'sep_conv_5x5' : lambda C, stride, affine: SepConv(C, C, 5, stride, 2, affine=affine), 'sep_conv_7x7' : lambda C, stride, affine: SepConv(C, C, 7, stride, 3, affine=affine), 'dil_conv_3x3' : lambda C, stride, affine: DilConv(C, C, 3, stride, 2, 2, affine=affine), 'dil_conv_5x5' : lambda C, stride, affine: DilConv(C, C, 5, stride, 4, 2, affine=affine), 'conv_7x1_1x7' : lambda C, stride, affine: nn.Sequential( nn.ReLU(inplace=False), nn.Conv2d(C, C, (1,7), stride=(1, stride), padding=(0, 3), bias=False), nn.Conv2d(C, C, (7,1), stride=(stride, 1), padding=(3, 0), bias=False), nn.BatchNorm2d(C, affine=affine) ), } class ReLUConvBN(nn.Module): def __init__(self, C_in, C_out, kernel_size, stride, padding, affine=True): super(ReLUConvBN, self).__init__() self.op = nn.Sequential( nn.ReLU(inplace=False), nn.Conv2d(C_in, C_out, kernel_size, stride=stride, padding=padding, bias=False), nn.BatchNorm2d(C_out, affine=affine) ) def forward(self, x): return self.op(x) class DilConv(nn.Module): def __init__(self, C_in, C_out, kernel_size, stride, padding, dilation, affine=True): super(DilConv, self).__init__() self.op = nn.Sequential( nn.ReLU(inplace=False), nn.Conv2d(C_in, C_in, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation, groups=C_in, bias=False), nn.Conv2d(C_in, C_out, kernel_size=1, padding=0, bias=False), nn.BatchNorm2d(C_out, affine=affine), ) def forward(self, x): return self.op(x) class SepConv(nn.Module): def __init__(self, C_in, C_out, kernel_size, stride, padding, affine=True): super(SepConv, self).__init__() self.op = nn.Sequential( nn.ReLU(inplace=False), nn.Conv2d(C_in, C_in, kernel_size=kernel_size, stride=stride, padding=padding, groups=C_in, bias=False), nn.Conv2d(C_in, C_in, kernel_size=1, padding=0, bias=False), nn.BatchNorm2d(C_in, affine=affine), nn.ReLU(inplace=False), nn.Conv2d(C_in, C_in, kernel_size=kernel_size, stride=1, padding=padding, groups=C_in, bias=False), nn.Conv2d(C_in, C_out, kernel_size=1, padding=0, bias=False), nn.BatchNorm2d(C_out, affine=affine), ) def forward(self, x): return self.op(x) class Identity(nn.Module): def __init__(self): super(Identity, self).__init__() def forward(self, x): return x class Zero(nn.Module): def __init__(self, stride): super(Zero, self).__init__() self.stride = stride def forward(self, x): if self.stride == 1: return x.mul(0.) return x[:,:,::self.stride,::self.stride].mul(0.) class FactorizedReduce(nn.Module): def __init__(self, C_in, C_out, affine=True): super(FactorizedReduce, self).__init__() assert C_out % 2 == 0 self.relu = nn.ReLU(inplace=False) self.conv_1 = nn.Conv2d(C_in, C_out // 2, 1, stride=2, padding=0, bias=False) self.conv_2 = nn.Conv2d(C_in, C_out // 2, 1, stride=2, padding=0, bias=False) self.bn = nn.BatchNorm2d(C_out, affine=affine) def forward(self, x): x = self.relu(x) out = torch.cat([self.conv_1(x), self.conv_2(x[:,:,1:,1:])], dim=1) out = self.bn(out) return out
3,717
34.075472
129
py
darts
darts-master/rnn/test.py
import argparse import os, sys import time import math import numpy as np import torch import torch.nn as nn import torch.backends.cudnn as cudnn import data import model from utils import batchify, get_batch, repackage_hidden, create_exp_dir, save_checkpoint parser = argparse.ArgumentParser(description='PyTorch PennTreeBank/WikiText2 Language Model') parser.add_argument('--data', type=str, default='../data/penn/', help='location of the data corpus') parser.add_argument('--emsize', type=int, default=850, help='size of word embeddings') parser.add_argument('--nhid', type=int, default=850, help='number of hidden units per layer') parser.add_argument('--nhidlast', type=int, default=850, help='number of hidden units for the last rnn layer') parser.add_argument('--lr', type=float, default=20, help='initial learning rate') parser.add_argument('--clip', type=float, default=0.25, help='gradient clipping') parser.add_argument('--epochs', type=int, default=8000, help='upper epoch limit') parser.add_argument('--batch_size', type=int, default=64, metavar='N', help='batch size') parser.add_argument('--bptt', type=int, default=35, help='sequence length') parser.add_argument('--dropout', type=float, default=0.75, help='dropout applied to layers (0 = no dropout)') parser.add_argument('--dropouth', type=float, default=0.3, help='dropout for rnn layers (0 = no dropout)') parser.add_argument('--dropouti', type=float, default=0.2, help='dropout for input embedding layers (0 = no dropout)') parser.add_argument('--dropoute', type=float, default=0.2, help='dropout to remove words from embedding layer (0 = no dropout)') parser.add_argument('--seed', type=int, default=1267, help='random seed') parser.add_argument('--nonmono', type=int, default=5, help='random seed') parser.add_argument('--cuda', action='store_false', help='use CUDA') parser.add_argument('--log-interval', type=int, default=200, metavar='N', help='report interval') parser.add_argument('--model_path', type=str, default='EXP/model.pt', help='path to load the pretrained model') parser.add_argument('--alpha', type=float, default=0, help='alpha L2 regularization on RNN activation (alpha = 0 means no regularization)') parser.add_argument('--beta', type=float, default=1e-3, help='beta slowness regularization applied on RNN activiation (beta = 0 means no regularization)') parser.add_argument('--wdecay', type=float, default=5e-7, help='weight decay applied to all weights') parser.add_argument('--continue_train', action='store_true', help='continue train from a checkpoint') parser.add_argument('--n_experts', type=int, default=1, help='number of experts') parser.add_argument('--max_seq_len_delta', type=int, default=20, help='max sequence length') parser.add_argument('--gpu', type=int, default=0, help='GPU device to use') args = parser.parse_args() def logging(s, print_=True, log_=True): print(s) # Set the random seed manually for reproducibility. np.random.seed(args.seed) torch.manual_seed(args.seed) if torch.cuda.is_available(): if not args.cuda: print("WARNING: You have a CUDA device, so you should probably run with --cuda") else: torch.cuda.set_device(args.gpu) cudnn.benchmark = True cudnn.enabled=True torch.cuda.manual_seed_all(args.seed) corpus = data.Corpus(args.data) test_batch_size = 1 test_data = batchify(corpus.test, test_batch_size, args) def evaluate(data_source, batch_size=10): # Turn on evaluation mode which disables dropout. model.eval() total_loss = 0 ntokens = len(corpus.dictionary) hidden = model.init_hidden(batch_size) for i in range(0, data_source.size(0) - 1, args.bptt): print(i, data_source.size(0)-1) data, targets = get_batch(data_source, i, args, evaluation=True) targets = targets.view(-1) log_prob, hidden = parallel_model(data, hidden) loss = nn.functional.nll_loss(log_prob.view(-1, log_prob.size(2)), targets).data total_loss += loss * len(data) hidden = repackage_hidden(hidden) return total_loss[0] / len(data_source) # Load the best saved model. model = torch.load(args.model_path) total_params = sum(x.data.nelement() for x in model.parameters()) logging('Args: {}'.format(args)) logging('Model total parameters: {}'.format(total_params)) parallel_model = model.cuda() # Run on test data. test_loss = evaluate(test_data, test_batch_size) logging('=' * 89) logging('| End of training | test loss {:5.2f} | test ppl {:8.2f}'.format( test_loss, math.exp(test_loss))) logging('=' * 89)
5,048
40.385246
118
py
darts
darts-master/rnn/architect.py
import torch import numpy as np import torch.nn as nn from torch.autograd import Variable def _concat(xs): return torch.cat([x.view(-1) for x in xs]) def _clip(grads, max_norm): total_norm = 0 for g in grads: param_norm = g.data.norm(2) total_norm += param_norm ** 2 total_norm = total_norm ** 0.5 clip_coef = max_norm / (total_norm + 1e-6) if clip_coef < 1: for g in grads: g.data.mul_(clip_coef) return clip_coef class Architect(object): def __init__(self, model, args): self.network_weight_decay = args.wdecay self.network_clip = args.clip self.model = model self.optimizer = torch.optim.Adam(self.model.arch_parameters(), lr=args.arch_lr, weight_decay=args.arch_wdecay) def _compute_unrolled_model(self, hidden, input, target, eta): loss, hidden_next = self.model._loss(hidden, input, target) theta = _concat(self.model.parameters()).data grads = torch.autograd.grad(loss, self.model.parameters()) clip_coef = _clip(grads, self.network_clip) dtheta = _concat(grads).data + self.network_weight_decay*theta unrolled_model = self._construct_model_from_theta(theta.sub(eta, dtheta)) return unrolled_model, clip_coef def step(self, hidden_train, input_train, target_train, hidden_valid, input_valid, target_valid, network_optimizer, unrolled): eta = network_optimizer.param_groups[0]['lr'] self.optimizer.zero_grad() if unrolled: hidden = self._backward_step_unrolled(hidden_train, input_train, target_train, hidden_valid, input_valid, target_valid, eta) else: hidden = self._backward_step(hidden_valid, input_valid, target_valid) self.optimizer.step() return hidden, None def _backward_step(self, hidden, input, target): loss, hidden_next = self.model._loss(hidden, input, target) loss.backward() return hidden_next def _backward_step_unrolled(self, hidden_train, input_train, target_train, hidden_valid, input_valid, target_valid, eta): unrolled_model, clip_coef = self._compute_unrolled_model(hidden_train, input_train, target_train, eta) unrolled_loss, hidden_next = unrolled_model._loss(hidden_valid, input_valid, target_valid) unrolled_loss.backward() dalpha = [v.grad for v in unrolled_model.arch_parameters()] dtheta = [v.grad for v in unrolled_model.parameters()] _clip(dtheta, self.network_clip) vector = [dt.data for dt in dtheta] implicit_grads = self._hessian_vector_product(vector, hidden_train, input_train, target_train, r=1e-2) for g, ig in zip(dalpha, implicit_grads): g.data.sub_(eta * clip_coef, ig.data) for v, g in zip(self.model.arch_parameters(), dalpha): if v.grad is None: v.grad = Variable(g.data) else: v.grad.data.copy_(g.data) return hidden_next def _construct_model_from_theta(self, theta): model_new = self.model.new() model_dict = self.model.state_dict() params, offset = {}, 0 for k, v in self.model.named_parameters(): v_length = np.prod(v.size()) params[k] = theta[offset: offset+v_length].view(v.size()) offset += v_length assert offset == len(theta) model_dict.update(params) model_new.load_state_dict(model_dict) return model_new.cuda() def _hessian_vector_product(self, vector, hidden, input, target, r=1e-2): R = r / _concat(vector).norm() for p, v in zip(self.model.parameters(), vector): p.data.add_(R, v) loss, _ = self.model._loss(hidden, input, target) grads_p = torch.autograd.grad(loss, self.model.arch_parameters()) for p, v in zip(self.model.parameters(), vector): p.data.sub_(2*R, v) loss, _ = self.model._loss(hidden, input, target) grads_n = torch.autograd.grad(loss, self.model.arch_parameters()) for p, v in zip(self.model.parameters(), vector): p.data.add_(R, v) return [(x-y).div_(2*R) for x, y in zip(grads_p, grads_n)]
4,003
34.122807
132
py
darts
darts-master/rnn/utils.py
import torch import torch.nn as nn import os, shutil import numpy as np from torch.autograd import Variable def repackage_hidden(h): if type(h) == Variable: return Variable(h.data) else: return tuple(repackage_hidden(v) for v in h) def batchify(data, bsz, args): nbatch = data.size(0) // bsz data = data.narrow(0, 0, nbatch * bsz) data = data.view(bsz, -1).t().contiguous() print(data.size()) if args.cuda: data = data.cuda() return data def get_batch(source, i, args, seq_len=None, evaluation=False): seq_len = min(seq_len if seq_len else args.bptt, len(source) - 1 - i) data = Variable(source[i:i+seq_len], volatile=evaluation) target = Variable(source[i+1:i+1+seq_len]) return data, target def create_exp_dir(path, scripts_to_save=None): if not os.path.exists(path): os.mkdir(path) print('Experiment dir : {}'.format(path)) if scripts_to_save is not None: os.mkdir(os.path.join(path, 'scripts')) for script in scripts_to_save: dst_file = os.path.join(path, 'scripts', os.path.basename(script)) shutil.copyfile(script, dst_file) def save_checkpoint(model, optimizer, epoch, path, finetune=False): if finetune: torch.save(model, os.path.join(path, 'finetune_model.pt')) torch.save(optimizer.state_dict(), os.path.join(path, 'finetune_optimizer.pt')) else: torch.save(model, os.path.join(path, 'model.pt')) torch.save(optimizer.state_dict(), os.path.join(path, 'optimizer.pt')) torch.save({'epoch': epoch+1}, os.path.join(path, 'misc.pt')) def embedded_dropout(embed, words, dropout=0.1, scale=None): if dropout: mask = embed.weight.data.new().resize_((embed.weight.size(0), 1)).bernoulli_(1 - dropout).expand_as(embed.weight) / (1 - dropout) mask = Variable(mask) masked_embed_weight = mask * embed.weight else: masked_embed_weight = embed.weight if scale: masked_embed_weight = scale.expand_as(masked_embed_weight) * masked_embed_weight padding_idx = embed.padding_idx if padding_idx is None: padding_idx = -1 X = embed._backend.Embedding.apply(words, masked_embed_weight, padding_idx, embed.max_norm, embed.norm_type, embed.scale_grad_by_freq, embed.sparse ) return X class LockedDropout(nn.Module): def __init__(self): super(LockedDropout, self).__init__() def forward(self, x, dropout=0.5): if not self.training or not dropout: return x m = x.data.new(1, x.size(1), x.size(2)).bernoulli_(1 - dropout) mask = Variable(m.div_(1 - dropout), requires_grad=False) mask = mask.expand_as(x) return mask * x def mask2d(B, D, keep_prob, cuda=True): m = torch.floor(torch.rand(B, D) + keep_prob) / keep_prob m = Variable(m, requires_grad=False) if cuda: m = m.cuda() return m
2,955
30.446809
137
py
darts
darts-master/rnn/model.py
import math import torch import torch.nn as nn import torch.nn.functional as F from genotypes import STEPS from utils import mask2d from utils import LockedDropout from utils import embedded_dropout from torch.autograd import Variable INITRANGE = 0.04 class DARTSCell(nn.Module): def __init__(self, ninp, nhid, dropouth, dropoutx, genotype): super(DARTSCell, self).__init__() self.nhid = nhid self.dropouth = dropouth self.dropoutx = dropoutx self.genotype = genotype # genotype is None when doing arch search steps = len(self.genotype.recurrent) if self.genotype is not None else STEPS self._W0 = nn.Parameter(torch.Tensor(ninp+nhid, 2*nhid).uniform_(-INITRANGE, INITRANGE)) self._Ws = nn.ParameterList([ nn.Parameter(torch.Tensor(nhid, 2*nhid).uniform_(-INITRANGE, INITRANGE)) for i in range(steps) ]) def forward(self, inputs, hidden): T, B = inputs.size(0), inputs.size(1) if self.training: x_mask = mask2d(B, inputs.size(2), keep_prob=1.-self.dropoutx) h_mask = mask2d(B, hidden.size(2), keep_prob=1.-self.dropouth) else: x_mask = h_mask = None hidden = hidden[0] hiddens = [] for t in range(T): hidden = self.cell(inputs[t], hidden, x_mask, h_mask) hiddens.append(hidden) hiddens = torch.stack(hiddens) return hiddens, hiddens[-1].unsqueeze(0) def _compute_init_state(self, x, h_prev, x_mask, h_mask): if self.training: xh_prev = torch.cat([x * x_mask, h_prev * h_mask], dim=-1) else: xh_prev = torch.cat([x, h_prev], dim=-1) c0, h0 = torch.split(xh_prev.mm(self._W0), self.nhid, dim=-1) c0 = c0.sigmoid() h0 = h0.tanh() s0 = h_prev + c0 * (h0-h_prev) return s0 def _get_activation(self, name): if name == 'tanh': f = F.tanh elif name == 'relu': f = F.relu elif name == 'sigmoid': f = F.sigmoid elif name == 'identity': f = lambda x: x else: raise NotImplementedError return f def cell(self, x, h_prev, x_mask, h_mask): s0 = self._compute_init_state(x, h_prev, x_mask, h_mask) states = [s0] for i, (name, pred) in enumerate(self.genotype.recurrent): s_prev = states[pred] if self.training: ch = (s_prev * h_mask).mm(self._Ws[i]) else: ch = s_prev.mm(self._Ws[i]) c, h = torch.split(ch, self.nhid, dim=-1) c = c.sigmoid() fn = self._get_activation(name) h = fn(h) s = s_prev + c * (h-s_prev) states += [s] output = torch.mean(torch.stack([states[i] for i in self.genotype.concat], -1), -1) return output class RNNModel(nn.Module): """Container module with an encoder, a recurrent module, and a decoder.""" def __init__(self, ntoken, ninp, nhid, nhidlast, dropout=0.5, dropouth=0.5, dropoutx=0.5, dropouti=0.5, dropoute=0.1, cell_cls=DARTSCell, genotype=None): super(RNNModel, self).__init__() self.lockdrop = LockedDropout() self.encoder = nn.Embedding(ntoken, ninp) assert ninp == nhid == nhidlast if cell_cls == DARTSCell: assert genotype is not None self.rnns = [cell_cls(ninp, nhid, dropouth, dropoutx, genotype)] else: assert genotype is None self.rnns = [cell_cls(ninp, nhid, dropouth, dropoutx)] self.rnns = torch.nn.ModuleList(self.rnns) self.decoder = nn.Linear(ninp, ntoken) self.decoder.weight = self.encoder.weight self.init_weights() self.ninp = ninp self.nhid = nhid self.nhidlast = nhidlast self.dropout = dropout self.dropouti = dropouti self.dropoute = dropoute self.ntoken = ntoken self.cell_cls = cell_cls def init_weights(self): self.encoder.weight.data.uniform_(-INITRANGE, INITRANGE) self.decoder.bias.data.fill_(0) self.decoder.weight.data.uniform_(-INITRANGE, INITRANGE) def forward(self, input, hidden, return_h=False): batch_size = input.size(1) emb = embedded_dropout(self.encoder, input, dropout=self.dropoute if self.training else 0) emb = self.lockdrop(emb, self.dropouti) raw_output = emb new_hidden = [] raw_outputs = [] outputs = [] for l, rnn in enumerate(self.rnns): current_input = raw_output raw_output, new_h = rnn(raw_output, hidden[l]) new_hidden.append(new_h) raw_outputs.append(raw_output) hidden = new_hidden output = self.lockdrop(raw_output, self.dropout) outputs.append(output) logit = self.decoder(output.view(-1, self.ninp)) log_prob = nn.functional.log_softmax(logit, dim=-1) model_output = log_prob model_output = model_output.view(-1, batch_size, self.ntoken) if return_h: return model_output, hidden, raw_outputs, outputs return model_output, hidden def init_hidden(self, bsz): weight = next(self.parameters()).data return [Variable(weight.new(1, bsz, self.nhid).zero_())]
5,148
30.981366
102
py
darts
darts-master/rnn/data.py
import os import torch from collections import Counter class Dictionary(object): def __init__(self): self.word2idx = {} self.idx2word = [] self.counter = Counter() self.total = 0 def add_word(self, word): if word not in self.word2idx: self.idx2word.append(word) self.word2idx[word] = len(self.idx2word) - 1 token_id = self.word2idx[word] self.counter[token_id] += 1 self.total += 1 return self.word2idx[word] def __len__(self): return len(self.idx2word) class Corpus(object): def __init__(self, path): self.dictionary = Dictionary() self.train = self.tokenize(os.path.join(path, 'train.txt')) self.valid = self.tokenize(os.path.join(path, 'valid.txt')) self.test = self.tokenize(os.path.join(path, 'test.txt')) def tokenize(self, path): """Tokenizes a text file.""" assert os.path.exists(path) # Add words to the dictionary with open(path, 'r', encoding='utf-8') as f: tokens = 0 for line in f: words = line.split() + ['<eos>'] tokens += len(words) for word in words: self.dictionary.add_word(word) # Tokenize file content with open(path, 'r', encoding='utf-8') as f: ids = torch.LongTensor(tokens) token = 0 for line in f: words = line.split() + ['<eos>'] for word in words: ids[token] = self.dictionary.word2idx[word] token += 1 return ids class SentCorpus(object): def __init__(self, path): self.dictionary = Dictionary() self.train = self.tokenize(os.path.join(path, 'train.txt')) self.valid = self.tokenize(os.path.join(path, 'valid.txt')) self.test = self.tokenize(os.path.join(path, 'test.txt')) def tokenize(self, path): """Tokenizes a text file.""" assert os.path.exists(path) # Add words to the dictionary with open(path, 'r', encoding='utf-8') as f: tokens = 0 for line in f: words = line.split() + ['<eos>'] tokens += len(words) for word in words: self.dictionary.add_word(word) # Tokenize file content sents = [] with open(path, 'r', encoding='utf-8') as f: for line in f: if not line: continue words = line.split() + ['<eos>'] sent = torch.LongTensor(len(words)) for i, word in enumerate(words): sent[i] = self.dictionary.word2idx[word] sents.append(sent) return sents class BatchSentLoader(object): def __init__(self, sents, batch_size, pad_id=0, cuda=False, volatile=False): self.sents = sents self.batch_size = batch_size self.sort_sents = sorted(sents, key=lambda x: x.size(0)) self.cuda = cuda self.volatile = volatile self.pad_id = pad_id def __next__(self): if self.idx >= len(self.sort_sents): raise StopIteration batch_size = min(self.batch_size, len(self.sort_sents)-self.idx) batch = self.sort_sents[self.idx:self.idx+batch_size] max_len = max([s.size(0) for s in batch]) tensor = torch.LongTensor(max_len, batch_size).fill_(self.pad_id) for i in range(len(batch)): s = batch[i] tensor[:s.size(0),i].copy_(s) if self.cuda: tensor = tensor.cuda() self.idx += batch_size return tensor next = __next__ def __iter__(self): self.idx = 0 return self if __name__ == '__main__': corpus = SentCorpus('../penn') loader = BatchSentLoader(corpus.test, 10) for i, d in enumerate(loader): print(i, d.size())
4,005
30.054264
80
py
darts
darts-master/rnn/model_search.py
import torch import torch.nn as nn import torch.nn.functional as F from genotypes import PRIMITIVES, STEPS, CONCAT, Genotype from torch.autograd import Variable from collections import namedtuple from model import DARTSCell, RNNModel class DARTSCellSearch(DARTSCell): def __init__(self, ninp, nhid, dropouth, dropoutx): super(DARTSCellSearch, self).__init__(ninp, nhid, dropouth, dropoutx, genotype=None) self.bn = nn.BatchNorm1d(nhid, affine=False) def cell(self, x, h_prev, x_mask, h_mask): s0 = self._compute_init_state(x, h_prev, x_mask, h_mask) s0 = self.bn(s0) probs = F.softmax(self.weights, dim=-1) offset = 0 states = s0.unsqueeze(0) for i in range(STEPS): if self.training: masked_states = states * h_mask.unsqueeze(0) else: masked_states = states ch = masked_states.view(-1, self.nhid).mm(self._Ws[i]).view(i+1, -1, 2*self.nhid) c, h = torch.split(ch, self.nhid, dim=-1) c = c.sigmoid() s = torch.zeros_like(s0) for k, name in enumerate(PRIMITIVES): if name == 'none': continue fn = self._get_activation(name) unweighted = states + c * (fn(h) - states) s += torch.sum(probs[offset:offset+i+1, k].unsqueeze(-1).unsqueeze(-1) * unweighted, dim=0) s = self.bn(s) states = torch.cat([states, s.unsqueeze(0)], 0) offset += i+1 output = torch.mean(states[-CONCAT:], dim=0) return output class RNNModelSearch(RNNModel): def __init__(self, *args): super(RNNModelSearch, self).__init__(*args, cell_cls=DARTSCellSearch, genotype=None) self._args = args self._initialize_arch_parameters() def new(self): model_new = RNNModelSearch(*self._args) for x, y in zip(model_new.arch_parameters(), self.arch_parameters()): x.data.copy_(y.data) return model_new def _initialize_arch_parameters(self): k = sum(i for i in range(1, STEPS+1)) weights_data = torch.randn(k, len(PRIMITIVES)).mul_(1e-3) self.weights = Variable(weights_data.cuda(), requires_grad=True) self._arch_parameters = [self.weights] for rnn in self.rnns: rnn.weights = self.weights def arch_parameters(self): return self._arch_parameters def _loss(self, hidden, input, target): log_prob, hidden_next = self(input, hidden, return_h=False) loss = nn.functional.nll_loss(log_prob.view(-1, log_prob.size(2)), target) return loss, hidden_next def genotype(self): def _parse(probs): gene = [] start = 0 for i in range(STEPS): end = start + i + 1 W = probs[start:end].copy() j = sorted(range(i + 1), key=lambda x: -max(W[x][k] for k in range(len(W[x])) if k != PRIMITIVES.index('none')))[0] k_best = None for k in range(len(W[j])): if k != PRIMITIVES.index('none'): if k_best is None or W[j][k] > W[j][k_best]: k_best = k gene.append((PRIMITIVES[k_best], j)) start = end return gene gene = _parse(F.softmax(self.weights, dim=-1).data.cpu().numpy()) genotype = Genotype(recurrent=gene, concat=range(STEPS+1)[-CONCAT:]) return genotype
3,278
32.804124
125
py
darts
darts-master/rnn/train_search.py
import argparse import os, sys, glob import time import math import numpy as np import torch import logging import torch.nn as nn import torch.nn.functional as F import torch.backends.cudnn as cudnn from architect import Architect import gc import data import model_search as model from utils import batchify, get_batch, repackage_hidden, create_exp_dir, save_checkpoint parser = argparse.ArgumentParser(description='PyTorch PennTreeBank/WikiText2 Language Model') parser.add_argument('--data', type=str, default='../data/penn/', help='location of the data corpus') parser.add_argument('--emsize', type=int, default=300, help='size of word embeddings') parser.add_argument('--nhid', type=int, default=300, help='number of hidden units per layer') parser.add_argument('--nhidlast', type=int, default=300, help='number of hidden units for the last rnn layer') parser.add_argument('--lr', type=float, default=20, help='initial learning rate') parser.add_argument('--clip', type=float, default=0.25, help='gradient clipping') parser.add_argument('--epochs', type=int, default=50, help='upper epoch limit') parser.add_argument('--batch_size', type=int, default=256, metavar='N', help='batch size') parser.add_argument('--bptt', type=int, default=35, help='sequence length') parser.add_argument('--dropout', type=float, default=0.75, help='dropout applied to layers (0 = no dropout)') parser.add_argument('--dropouth', type=float, default=0.25, help='dropout for hidden nodes in rnn layers (0 = no dropout)') parser.add_argument('--dropoutx', type=float, default=0.75, help='dropout for input nodes in rnn layers (0 = no dropout)') parser.add_argument('--dropouti', type=float, default=0.2, help='dropout for input embedding layers (0 = no dropout)') parser.add_argument('--dropoute', type=float, default=0, help='dropout to remove words from embedding layer (0 = no dropout)') parser.add_argument('--seed', type=int, default=3, help='random seed') parser.add_argument('--nonmono', type=int, default=5, help='random seed') parser.add_argument('--cuda', action='store_false', help='use CUDA') parser.add_argument('--log-interval', type=int, default=50, metavar='N', help='report interval') parser.add_argument('--save', type=str, default='EXP', help='path to save the final model') parser.add_argument('--alpha', type=float, default=0, help='alpha L2 regularization on RNN activation (alpha = 0 means no regularization)') parser.add_argument('--beta', type=float, default=1e-3, help='beta slowness regularization applied on RNN activiation (beta = 0 means no regularization)') parser.add_argument('--wdecay', type=float, default=5e-7, help='weight decay applied to all weights') parser.add_argument('--continue_train', action='store_true', help='continue train from a checkpoint') parser.add_argument('--small_batch_size', type=int, default=-1, help='the batch size for computation. batch_size should be divisible by small_batch_size.\ In our implementation, we compute gradients with small_batch_size multiple times, and accumulate the gradients\ until batch_size is reached. An update step is then performed.') parser.add_argument('--max_seq_len_delta', type=int, default=20, help='max sequence length') parser.add_argument('--single_gpu', default=True, action='store_false', help='use single GPU') parser.add_argument('--gpu', type=int, default=0, help='GPU device to use') parser.add_argument('--unrolled', action='store_true', default=False, help='use one-step unrolled validation loss') parser.add_argument('--arch_wdecay', type=float, default=1e-3, help='weight decay for the architecture encoding alpha') parser.add_argument('--arch_lr', type=float, default=3e-3, help='learning rate for the architecture encoding alpha') args = parser.parse_args() if args.nhidlast < 0: args.nhidlast = args.emsize if args.small_batch_size < 0: args.small_batch_size = args.batch_size if not args.continue_train: args.save = 'search-{}-{}'.format(args.save, time.strftime("%Y%m%d-%H%M%S")) create_exp_dir(args.save, scripts_to_save=glob.glob('*.py')) log_format = '%(asctime)s %(message)s' logging.basicConfig(stream=sys.stdout, level=logging.INFO, format=log_format, datefmt='%m/%d %I:%M:%S %p') fh = logging.FileHandler(os.path.join(args.save, 'log.txt')) fh.setFormatter(logging.Formatter(log_format)) logging.getLogger().addHandler(fh) # Set the random seed manually for reproducibility. np.random.seed(args.seed) torch.manual_seed(args.seed) if torch.cuda.is_available(): if not args.cuda: print("WARNING: You have a CUDA device, so you should probably run with --cuda") else: torch.cuda.set_device(args.gpu) cudnn.benchmark = True cudnn.enabled=True torch.cuda.manual_seed_all(args.seed) corpus = data.Corpus(args.data) eval_batch_size = 10 test_batch_size = 1 train_data = batchify(corpus.train, args.batch_size, args) search_data = batchify(corpus.valid, args.batch_size, args) val_data = batchify(corpus.valid, eval_batch_size, args) test_data = batchify(corpus.test, test_batch_size, args) ntokens = len(corpus.dictionary) if args.continue_train: model = torch.load(os.path.join(args.save, 'model.pt')) else: model = model.RNNModelSearch(ntokens, args.emsize, args.nhid, args.nhidlast, args.dropout, args.dropouth, args.dropoutx, args.dropouti, args.dropoute) size = 0 for p in model.parameters(): size += p.nelement() logging.info('param size: {}'.format(size)) logging.info('initial genotype:') logging.info(model.genotype()) if args.cuda: if args.single_gpu: parallel_model = model.cuda() else: parallel_model = nn.DataParallel(model, dim=1).cuda() else: parallel_model = model architect = Architect(parallel_model, args) total_params = sum(x.data.nelement() for x in model.parameters()) logging.info('Args: {}'.format(args)) logging.info('Model total parameters: {}'.format(total_params)) def evaluate(data_source, batch_size=10): # Turn on evaluation mode which disables dropout. model.eval() total_loss = 0 ntokens = len(corpus.dictionary) hidden = model.init_hidden(batch_size) for i in range(0, data_source.size(0) - 1, args.bptt): data, targets = get_batch(data_source, i, args, evaluation=True) targets = targets.view(-1) log_prob, hidden = parallel_model(data, hidden) loss = nn.functional.nll_loss(log_prob.view(-1, log_prob.size(2)), targets).data total_loss += loss * len(data) hidden = repackage_hidden(hidden) return total_loss[0] / len(data_source) def train(): assert args.batch_size % args.small_batch_size == 0, 'batch_size must be divisible by small_batch_size' # Turn on training mode which enables dropout. total_loss = 0 start_time = time.time() ntokens = len(corpus.dictionary) hidden = [model.init_hidden(args.small_batch_size) for _ in range(args.batch_size // args.small_batch_size)] hidden_valid = [model.init_hidden(args.small_batch_size) for _ in range(args.batch_size // args.small_batch_size)] batch, i = 0, 0 while i < train_data.size(0) - 1 - 1: bptt = args.bptt if np.random.random() < 0.95 else args.bptt / 2. # Prevent excessively small or negative sequence lengths # seq_len = max(5, int(np.random.normal(bptt, 5))) # # There's a very small chance that it could select a very long sequence length resulting in OOM # seq_len = min(seq_len, args.bptt + args.max_seq_len_delta) seq_len = int(bptt) lr2 = optimizer.param_groups[0]['lr'] optimizer.param_groups[0]['lr'] = lr2 * seq_len / args.bptt model.train() data_valid, targets_valid = get_batch(search_data, i % (search_data.size(0) - 1), args) data, targets = get_batch(train_data, i, args, seq_len=seq_len) optimizer.zero_grad() start, end, s_id = 0, args.small_batch_size, 0 while start < args.batch_size: cur_data, cur_targets = data[:, start: end], targets[:, start: end].contiguous().view(-1) cur_data_valid, cur_targets_valid = data_valid[:, start: end], targets_valid[:, start: end].contiguous().view(-1) # Starting each batch, we detach the hidden state from how it was previously produced. # If we didn't, the model would try backpropagating all the way to start of the dataset. hidden[s_id] = repackage_hidden(hidden[s_id]) hidden_valid[s_id] = repackage_hidden(hidden_valid[s_id]) hidden_valid[s_id], grad_norm = architect.step( hidden[s_id], cur_data, cur_targets, hidden_valid[s_id], cur_data_valid, cur_targets_valid, optimizer, args.unrolled) # assuming small_batch_size = batch_size so we don't accumulate gradients optimizer.zero_grad() hidden[s_id] = repackage_hidden(hidden[s_id]) log_prob, hidden[s_id], rnn_hs, dropped_rnn_hs = parallel_model(cur_data, hidden[s_id], return_h=True) raw_loss = nn.functional.nll_loss(log_prob.view(-1, log_prob.size(2)), cur_targets) loss = raw_loss # Activiation Regularization if args.alpha > 0: loss = loss + sum(args.alpha * dropped_rnn_h.pow(2).mean() for dropped_rnn_h in dropped_rnn_hs[-1:]) # Temporal Activation Regularization (slowness) loss = loss + sum(args.beta * (rnn_h[1:] - rnn_h[:-1]).pow(2).mean() for rnn_h in rnn_hs[-1:]) loss *= args.small_batch_size / args.batch_size total_loss += raw_loss.data * args.small_batch_size / args.batch_size loss.backward() s_id += 1 start = end end = start + args.small_batch_size gc.collect() # `clip_grad_norm` helps prevent the exploding gradient problem in RNNs. torch.nn.utils.clip_grad_norm(model.parameters(), args.clip) optimizer.step() # total_loss += raw_loss.data optimizer.param_groups[0]['lr'] = lr2 if batch % args.log_interval == 0 and batch > 0: logging.info(parallel_model.genotype()) print(F.softmax(parallel_model.weights, dim=-1)) cur_loss = total_loss[0] / args.log_interval elapsed = time.time() - start_time logging.info('| epoch {:3d} | {:5d}/{:5d} batches | lr {:02.2f} | ms/batch {:5.2f} | ' 'loss {:5.2f} | ppl {:8.2f}'.format( epoch, batch, len(train_data) // args.bptt, optimizer.param_groups[0]['lr'], elapsed * 1000 / args.log_interval, cur_loss, math.exp(cur_loss))) total_loss = 0 start_time = time.time() batch += 1 i += seq_len # Loop over epochs. lr = args.lr best_val_loss = [] stored_loss = 100000000 if args.continue_train: optimizer_state = torch.load(os.path.join(args.save, 'optimizer.pt')) if 't0' in optimizer_state['param_groups'][0]: optimizer = torch.optim.ASGD(model.parameters(), lr=args.lr, t0=0, lambd=0., weight_decay=args.wdecay) else: optimizer = torch.optim.SGD(model.parameters(), lr=args.lr, weight_decay=args.wdecay) optimizer.load_state_dict(optimizer_state) else: optimizer = torch.optim.SGD(model.parameters(), lr=args.lr, weight_decay=args.wdecay) for epoch in range(1, args.epochs+1): epoch_start_time = time.time() train() val_loss = evaluate(val_data, eval_batch_size) logging.info('-' * 89) logging.info('| end of epoch {:3d} | time: {:5.2f}s | valid loss {:5.2f} | ' 'valid ppl {:8.2f}'.format(epoch, (time.time() - epoch_start_time), val_loss, math.exp(val_loss))) logging.info('-' * 89) if val_loss < stored_loss: save_checkpoint(model, optimizer, epoch, args.save) logging.info('Saving Normal!') stored_loss = val_loss best_val_loss.append(val_loss)
12,639
43.041812
132
py
darts
darts-master/rnn/train.py
import os import gc import sys import glob import time import math import numpy as np import torch import torch.nn as nn import logging import argparse import genotypes import torch.nn.functional as F import torch.backends.cudnn as cudnn import data import model from torch.autograd import Variable from utils import batchify, get_batch, repackage_hidden, create_exp_dir, save_checkpoint parser = argparse.ArgumentParser(description='PyTorch PennTreeBank/WikiText2 Language Model') parser.add_argument('--data', type=str, default='../data/penn/', help='location of the data corpus') parser.add_argument('--emsize', type=int, default=850, help='size of word embeddings') parser.add_argument('--nhid', type=int, default=850, help='number of hidden units per layer') parser.add_argument('--nhidlast', type=int, default=850, help='number of hidden units for the last rnn layer') parser.add_argument('--lr', type=float, default=20, help='initial learning rate') parser.add_argument('--clip', type=float, default=0.25, help='gradient clipping') parser.add_argument('--epochs', type=int, default=8000, help='upper epoch limit') parser.add_argument('--batch_size', type=int, default=64, metavar='N', help='batch size') parser.add_argument('--bptt', type=int, default=35, help='sequence length') parser.add_argument('--dropout', type=float, default=0.75, help='dropout applied to layers (0 = no dropout)') parser.add_argument('--dropouth', type=float, default=0.25, help='dropout for hidden nodes in rnn layers (0 = no dropout)') parser.add_argument('--dropoutx', type=float, default=0.75, help='dropout for input nodes rnn layers (0 = no dropout)') parser.add_argument('--dropouti', type=float, default=0.2, help='dropout for input embedding layers (0 = no dropout)') parser.add_argument('--dropoute', type=float, default=0.1, help='dropout to remove words from embedding layer (0 = no dropout)') parser.add_argument('--seed', type=int, default=1267, help='random seed') parser.add_argument('--nonmono', type=int, default=5, help='random seed') parser.add_argument('--cuda', action='store_false', help='use CUDA') parser.add_argument('--log-interval', type=int, default=200, metavar='N', help='report interval') parser.add_argument('--save', type=str, default='EXP', help='path to save the final model') parser.add_argument('--alpha', type=float, default=0, help='alpha L2 regularization on RNN activation (alpha = 0 means no regularization)') parser.add_argument('--beta', type=float, default=1e-3, help='beta slowness regularization applied on RNN activiation (beta = 0 means no regularization)') parser.add_argument('--wdecay', type=float, default=8e-7, help='weight decay applied to all weights') parser.add_argument('--continue_train', action='store_true', help='continue train from a checkpoint') parser.add_argument('--small_batch_size', type=int, default=-1, help='the batch size for computation. batch_size should be divisible by small_batch_size.\ In our implementation, we compute gradients with small_batch_size multiple times, and accumulate the gradients\ until batch_size is reached. An update step is then performed.') parser.add_argument('--max_seq_len_delta', type=int, default=20, help='max sequence length') parser.add_argument('--single_gpu', default=True, action='store_false', help='use single GPU') parser.add_argument('--gpu', type=int, default=0, help='GPU device to use') parser.add_argument('--arch', type=str, default='DARTS', help='which architecture to use') args = parser.parse_args() if args.nhidlast < 0: args.nhidlast = args.emsize if args.small_batch_size < 0: args.small_batch_size = args.batch_size if not args.continue_train: args.save = 'eval-{}-{}'.format(args.save, time.strftime("%Y%m%d-%H%M%S")) create_exp_dir(args.save, scripts_to_save=glob.glob('*.py')) log_format = '%(asctime)s %(message)s' logging.basicConfig(stream=sys.stdout, level=logging.INFO, format=log_format, datefmt='%m/%d %I:%M:%S %p') fh = logging.FileHandler(os.path.join(args.save, 'log.txt')) fh.setFormatter(logging.Formatter(log_format)) logging.getLogger().addHandler(fh) # Set the random seed manually for reproducibility. np.random.seed(args.seed) torch.manual_seed(args.seed) if torch.cuda.is_available(): if not args.cuda: print("WARNING: You have a CUDA device, so you should probably run with --cuda") else: torch.cuda.set_device(args.gpu) cudnn.benchmark = True cudnn.enabled=True torch.cuda.manual_seed_all(args.seed) corpus = data.Corpus(args.data) eval_batch_size = 10 test_batch_size = 1 train_data = batchify(corpus.train, args.batch_size, args) val_data = batchify(corpus.valid, eval_batch_size, args) test_data = batchify(corpus.test, test_batch_size, args) ntokens = len(corpus.dictionary) if args.continue_train: model = torch.load(os.path.join(args.save, 'model.pt')) else: genotype = eval("genotypes.%s" % args.arch) model = model.RNNModel(ntokens, args.emsize, args.nhid, args.nhidlast, args.dropout, args.dropouth, args.dropoutx, args.dropouti, args.dropoute, cell_cls=model.DARTSCell, genotype=genotype) if args.cuda: if args.single_gpu: parallel_model = model.cuda() else: parallel_model = nn.DataParallel(model, dim=1).cuda() else: parallel_model = model total_params = sum(x.data.nelement() for x in model.parameters()) logging.info('Args: {}'.format(args)) logging.info('Model total parameters: {}'.format(total_params)) logging.info('Genotype: {}'.format(genotype)) def evaluate(data_source, batch_size=10): # Turn on evaluation mode which disables dropout. model.eval() total_loss = 0 ntokens = len(corpus.dictionary) hidden = model.init_hidden(batch_size) for i in range(0, data_source.size(0) - 1, args.bptt): data, targets = get_batch(data_source, i, args, evaluation=True) targets = targets.view(-1) log_prob, hidden = parallel_model(data, hidden) loss = nn.functional.nll_loss(log_prob.view(-1, log_prob.size(2)), targets).data total_loss += loss * len(data) hidden = repackage_hidden(hidden) return total_loss[0] / len(data_source) def train(): assert args.batch_size % args.small_batch_size == 0, 'batch_size must be divisible by small_batch_size' # Turn on training mode which enables dropout. total_loss = 0 start_time = time.time() ntokens = len(corpus.dictionary) hidden = [model.init_hidden(args.small_batch_size) for _ in range(args.batch_size // args.small_batch_size)] batch, i = 0, 0 while i < train_data.size(0) - 1 - 1: bptt = args.bptt if np.random.random() < 0.95 else args.bptt / 2. # Prevent excessively small or negative sequence lengths seq_len = max(5, int(np.random.normal(bptt, 5))) # There's a very small chance that it could select a very long sequence length resulting in OOM seq_len = min(seq_len, args.bptt + args.max_seq_len_delta) lr2 = optimizer.param_groups[0]['lr'] optimizer.param_groups[0]['lr'] = lr2 * seq_len / args.bptt model.train() data, targets = get_batch(train_data, i, args, seq_len=seq_len) optimizer.zero_grad() start, end, s_id = 0, args.small_batch_size, 0 while start < args.batch_size: cur_data, cur_targets = data[:, start: end], targets[:, start: end].contiguous().view(-1) # Starting each batch, we detach the hidden state from how it was previously produced. # If we didn't, the model would try backpropagating all the way to start of the dataset. hidden[s_id] = repackage_hidden(hidden[s_id]) log_prob, hidden[s_id], rnn_hs, dropped_rnn_hs = parallel_model(cur_data, hidden[s_id], return_h=True) raw_loss = nn.functional.nll_loss(log_prob.view(-1, log_prob.size(2)), cur_targets) loss = raw_loss # Activiation Regularization if args.alpha > 0: loss = loss + sum(args.alpha * dropped_rnn_h.pow(2).mean() for dropped_rnn_h in dropped_rnn_hs[-1:]) # Temporal Activation Regularization (slowness) loss = loss + sum(args.beta * (rnn_h[1:] - rnn_h[:-1]).pow(2).mean() for rnn_h in rnn_hs[-1:]) loss *= args.small_batch_size / args.batch_size total_loss += raw_loss.data * args.small_batch_size / args.batch_size loss.backward() s_id += 1 start = end end = start + args.small_batch_size gc.collect() # `clip_grad_norm` helps prevent the exploding gradient problem in RNNs. torch.nn.utils.clip_grad_norm(model.parameters(), args.clip) optimizer.step() # total_loss += raw_loss.data optimizer.param_groups[0]['lr'] = lr2 if np.isnan(total_loss[0]): raise if batch % args.log_interval == 0 and batch > 0: cur_loss = total_loss[0] / args.log_interval elapsed = time.time() - start_time logging.info('| epoch {:3d} | {:5d}/{:5d} batches | lr {:02.2f} | ms/batch {:5.2f} | ' 'loss {:5.2f} | ppl {:8.2f}'.format( epoch, batch, len(train_data) // args.bptt, optimizer.param_groups[0]['lr'], elapsed * 1000 / args.log_interval, cur_loss, math.exp(cur_loss))) total_loss = 0 start_time = time.time() batch += 1 i += seq_len # Loop over epochs. lr = args.lr best_val_loss = [] stored_loss = 100000000 # At any point you can hit Ctrl + C to break out of training early. try: if args.continue_train: optimizer_state = torch.load(os.path.join(args.save, 'optimizer.pt')) if 't0' in optimizer_state['param_groups'][0]: optimizer = torch.optim.ASGD(model.parameters(), lr=args.lr, t0=0, lambd=0., weight_decay=args.wdecay) else: optimizer = torch.optim.SGD(model.parameters(), lr=args.lr, weight_decay=args.wdecay) optimizer.load_state_dict(optimizer_state) else: optimizer = torch.optim.SGD(model.parameters(), lr=args.lr, weight_decay=args.wdecay) epoch = 1 while epoch < args.epochs + 1: epoch_start_time = time.time() try: train() except: logging.info('rolling back to the previous best model ...') model = torch.load(os.path.join(args.save, 'model.pt')) parallel_model = model.cuda() optimizer_state = torch.load(os.path.join(args.save, 'optimizer.pt')) if 't0' in optimizer_state['param_groups'][0]: optimizer = torch.optim.ASGD(model.parameters(), lr=args.lr, t0=0, lambd=0., weight_decay=args.wdecay) else: optimizer = torch.optim.SGD(model.parameters(), lr=args.lr, weight_decay=args.wdecay) optimizer.load_state_dict(optimizer_state) epoch = torch.load(os.path.join(args.save, 'misc.pt'))['epoch'] continue if 't0' in optimizer.param_groups[0]: tmp = {} for prm in model.parameters(): tmp[prm] = prm.data.clone() prm.data = optimizer.state[prm]['ax'].clone() val_loss2 = evaluate(val_data) logging.info('-' * 89) logging.info('| end of epoch {:3d} | time: {:5.2f}s | valid loss {:5.2f} | ' 'valid ppl {:8.2f}'.format(epoch, (time.time() - epoch_start_time), val_loss2, math.exp(val_loss2))) logging.info('-' * 89) if val_loss2 < stored_loss: save_checkpoint(model, optimizer, epoch, args.save) logging.info('Saving Averaged!') stored_loss = val_loss2 for prm in model.parameters(): prm.data = tmp[prm].clone() else: val_loss = evaluate(val_data, eval_batch_size) logging.info('-' * 89) logging.info('| end of epoch {:3d} | time: {:5.2f}s | valid loss {:5.2f} | ' 'valid ppl {:8.2f}'.format(epoch, (time.time() - epoch_start_time), val_loss, math.exp(val_loss))) logging.info('-' * 89) if val_loss < stored_loss: save_checkpoint(model, optimizer, epoch, args.save) logging.info('Saving Normal!') stored_loss = val_loss if 't0' not in optimizer.param_groups[0] and (len(best_val_loss)>args.nonmono and val_loss > min(best_val_loss[:-args.nonmono])): logging.info('Switching!') optimizer = torch.optim.ASGD(model.parameters(), lr=args.lr, t0=0, lambd=0., weight_decay=args.wdecay) best_val_loss.append(val_loss) epoch += 1 except KeyboardInterrupt: logging.info('-' * 89) logging.info('Exiting from training early') # Load the best saved model. model = torch.load(os.path.join(args.save, 'model.pt')) parallel_model = model.cuda() # Run on test data. test_loss = evaluate(test_data, test_batch_size) logging.info('=' * 89) logging.info('| End of training | test loss {:5.2f} | test ppl {:8.2f}'.format( test_loss, math.exp(test_loss))) logging.info('=' * 89)
13,900
42.037152
141
py
darts
darts-master/rnn/visualize.py
import sys import genotypes from graphviz import Digraph def plot(genotype, filename): g = Digraph( format='pdf', edge_attr=dict(fontsize='20', fontname="times"), node_attr=dict(style='filled', shape='rect', align='center', fontsize='20', height='0.5', width='0.5', penwidth='2', fontname="times"), engine='dot') g.body.extend(['rankdir=LR']) g.node("x_{t}", fillcolor='darkseagreen2') g.node("h_{t-1}", fillcolor='darkseagreen2') g.node("0", fillcolor='lightblue') g.edge("x_{t}", "0", fillcolor="gray") g.edge("h_{t-1}", "0", fillcolor="gray") steps = len(genotype) for i in range(1, steps + 1): g.node(str(i), fillcolor='lightblue') for i, (op, j) in enumerate(genotype): g.edge(str(j), str(i + 1), label=op, fillcolor="gray") g.node("h_{t}", fillcolor='palegoldenrod') for i in range(1, steps + 1): g.edge(str(i), "h_{t}", fillcolor="gray") g.render(filename, view=True) if __name__ == '__main__': if len(sys.argv) != 2: print("usage:\n python {} ARCH_NAME".format(sys.argv[0])) sys.exit(1) genotype_name = sys.argv[1] try: genotype = eval('genotypes.{}'.format(genotype_name)) except AttributeError: print("{} is not specified in genotypes.py".format(genotype_name)) sys.exit(1) plot(genotype.recurrent, "recurrent")
1,327
26.666667
141
py
darts
darts-master/rnn/genotypes.py
from collections import namedtuple Genotype = namedtuple('Genotype', 'recurrent concat') PRIMITIVES = [ 'none', 'tanh', 'relu', 'sigmoid', 'identity' ] STEPS = 8 CONCAT = 8 ENAS = Genotype( recurrent = [ ('tanh', 0), ('tanh', 1), ('relu', 1), ('tanh', 3), ('tanh', 3), ('relu', 3), ('relu', 4), ('relu', 7), ('relu', 8), ('relu', 8), ('relu', 8), ], concat = [2, 5, 6, 9, 10, 11] ) DARTS_V1 = Genotype(recurrent=[('relu', 0), ('relu', 1), ('tanh', 2), ('relu', 3), ('relu', 4), ('identity', 1), ('relu', 5), ('relu', 1)], concat=range(1, 9)) DARTS_V2 = Genotype(recurrent=[('sigmoid', 0), ('relu', 1), ('relu', 1), ('identity', 1), ('tanh', 2), ('sigmoid', 5), ('tanh', 3), ('relu', 5)], concat=range(1, 9)) DARTS = DARTS_V2
851
22.027027
165
py
MaskTextSpotterV3
MaskTextSpotterV3-master/setup.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. #!/usr/bin/env python import glob import os import torch from setuptools import find_packages from setuptools import setup from torch.utils.cpp_extension import CUDA_HOME from torch.utils.cpp_extension import CppExtension from torch.utils.cpp_extension import CUDAExtension requirements = ["torch", "torchvision"] def get_extensions(): this_dir = os.path.dirname(os.path.abspath(__file__)) extensions_dir = os.path.join(this_dir, "maskrcnn_benchmark", "csrc") main_file = glob.glob(os.path.join(extensions_dir, "*.cpp")) source_cpu = glob.glob(os.path.join(extensions_dir, "cpu", "*.cpp")) source_cuda = glob.glob(os.path.join(extensions_dir, "cuda", "*.cu")) sources = main_file + source_cpu extension = CppExtension extra_compile_args = {"cxx": []} define_macros = [] if torch.cuda.is_available() and CUDA_HOME is not None: # if True: extension = CUDAExtension sources += source_cuda define_macros += [("WITH_CUDA", None)] extra_compile_args["nvcc"] = [ "-DCUDA_HAS_FP16=1", "-D__CUDA_NO_HALF_OPERATORS__", "-D__CUDA_NO_HALF_CONVERSIONS__", "-D__CUDA_NO_HALF2_OPERATORS__", ] sources = [os.path.join(extensions_dir, s) for s in sources] include_dirs = [extensions_dir] ext_modules = [ extension( "maskrcnn_benchmark._C", sources, include_dirs=include_dirs, define_macros=define_macros, extra_compile_args=extra_compile_args, ) ] return ext_modules setup( name="maskrcnn_benchmark", version="0.1", author="fmassa", url="https://github.com/facebookresearch/maskrnn-benchmark", description="object detection in pytorch", packages=find_packages(exclude=("configs", "examples", "test",)), # install_requires=requirements, ext_modules=get_extensions(), cmdclass={"build_ext": torch.utils.cpp_extension.BuildExtension}, )
2,068
28.140845
73
py
MaskTextSpotterV3
MaskTextSpotterV3-master/tools/convert_dataset.py
import os import numpy as np import cv2 from shapely.geometry import box, Polygon from shapely import affinity import math def _rect2quad(boxes): x_min, y_min, x_max, y_max = boxes[:, 0].reshape((-1, 1)), boxes[:, 1].reshape((-1, 1)), boxes[:, 2].reshape((-1, 1)), boxes[:, 3].reshape((-1, 1)) return np.hstack((x_min, y_min, x_max, y_min, x_max, y_max, x_min, y_max)) def _quad2rect(boxes): ## only support rectangle return np.hstack((boxes[:, 0].reshape((-1, 1)), boxes[:, 1].reshape((-1, 1)), boxes[:, 4].reshape((-1, 1)), boxes[:, 5].reshape((-1, 1)))) def _quad2minrect(boxes): ## trans a quad(N*4) to a rectangle(N*4) which has miniual area to cover it return np.hstack((boxes[:, ::2].min(axis=1).reshape((-1, 1)), boxes[:, 1::2].min(axis=1).reshape((-1, 1)), boxes[:, ::2].max(axis=1).reshape((-1, 1)), boxes[:, 1::2].max(axis=1).reshape((-1, 1)))) def _quad2boxlist(boxes): res = [] for i in range(boxes.shape[0]): res.append([[boxes[i][0], boxes[i][1]], [boxes[i][2], boxes[i][3]], [boxes[i][4], boxes[i][5]], [boxes[i][6], boxes[i][7]]]) return res def _boxlist2quads(boxlist): res = np.zeros((len(boxlist), 8)) for i, box in enumerate(boxlist): # print(box) res[i] = np.array([box[0][0], box[0][1], box[1][0], box[1][1], box[2][0], box[2][1], box[3][0], box[3][1]]) return res def _rotate_image(im, polygons, angle): new_polygons = polygons ## rotate image first height, width, _ = im.shape ## get the minimal rect to cover the rotated image img_box = np.array([[0, 0, width, 0, width, height, 0, height]]) rotated_img_box = _quad2minrect(_rotate_polygons(img_box, -1*angle, (width/2, height/2))) r_height = int(max(rotated_img_box[0][3], rotated_img_box[0][1]) - min(rotated_img_box[0][3], rotated_img_box[0][1])) r_width = int(max(rotated_img_box[0][2], rotated_img_box[0][0]) - min(rotated_img_box[0][2], rotated_img_box[0][0])) r_height_padding = max(r_height, height) r_width_padding = max(r_width, width) ## padding im im_padding = np.zeros((r_height_padding, r_width_padding, 3)) start_h, start_w = int((r_height_padding - height)/2.0), int((r_width_padding - width)/2.0) # start_h = max(start_h, 0) # start_w = max(start_w, 0) end_h, end_w = start_h + height, start_w + width # print(start_h, end_h, start_w, end_w, im.shape) im_padding[start_h:end_h, start_w:end_w, :] = im M = cv2.getRotationMatrix2D((r_width/2, r_height/2), angle, 1) im = cv2.warpAffine(im_padding, M, (r_width, r_height)) ## polygons new_polygons = _rotate_segms(polygons, -1*angle, (r_width/2, r_height/2), start_h, start_w) return im, new_polygons def _rotate_polygons(polygons, angle, r_c): ## polygons: N*8 ## r_x: rotate center x ## r_y: rotate center y ## angle: -15~15 poly_list = _quad2boxlist(polygons) rotate_boxes_list = [] for poly in poly_list: box = Polygon(poly) rbox = affinity.rotate(box, angle, r_c) if len(list(rbox.exterior.coords))<5: print(poly) print(rbox) # assert(len(list(rbox.exterior.coords))>=5) rotate_boxes_list.append(rbox.boundary.coords[:-1]) res = _boxlist2quads(rotate_boxes_list) return res def _rotate_segms(polygons, angle, r_c, start_h, start_w): ## polygons: N*8 ## r_x: rotate center x ## r_y: rotate center y ## angle: -15~15 poly_list=[] for polygon in polygons: tmp=[] for i in range(int(len(polygon) / 2)): tmp.append([polygon[2*i] + start_w, polygon[2*i+1] + start_h]) poly_list.append(tmp) rotate_boxes_list = [] for poly in poly_list: box = Polygon(poly) rbox = affinity.rotate(box, angle, r_c) if len(list(rbox.exterior.coords))<5: print(poly) print(rbox) rotate_boxes_list.append(rbox.boundary.coords[:-1]) res = [] for i, box in enumerate(rotate_boxes_list): tmp = [] for point in box: tmp.append(point[0]) tmp.append(point[1]) res.append([tmp]) return res def _read_gt(gt_path): polygons = [] words = [] with open(gt_path, 'r') as fid: lines = fid.readlines() for line in lines: line = line.strip() polygon = line.split(',')[:8] word = line.split(',')[8] polygon = [float(x) for x in polygon] polygons.append(polygon) words.append(word) return polygons, words def format_new_gt(polygons, words, new_gt_path): with open(new_gt_path, 'wt') as fid: for polygon, word in zip(polygons, words): # print(polygon) polygon = [str(int(x)) for x in polygon[0]] # polygon = [str(int(x)) for x in polygon] line = ','.join(polygon) + ',' + word # print(line) fid.write(line+'\n') def visu_gt(img, polygons, visu_path): for polygon in polygons: pts = np.array(polygon, np.int32) pts = pts.reshape((-1,1,2)) cv2.polylines(img,[pts],True,(0,255,255)) cv2.imwrite(visu_path, img) img_dir = '../datasets/icdar2013/test_images' gt_dir = '../datasets/icdar2013/test_gts' angle = 45 new_img_dir = '../datasets/icdar2013/rotated_test_images'+'_'+str(angle) new_gt_dir = '../datasets/icdar2013/rotated_test_gts'+'_'+str(angle) if not os.path.isdir(new_img_dir): os.mkdir(new_img_dir) if not os.path.isdir(new_gt_dir): os.mkdir(new_gt_dir) visu_dir = '../output/visu/' for i in range(233): img_name = 'img_' + str(i+1) + '.jpg' img_path = os.path.join(img_dir, img_name) img = cv2.imread(img_path) gt_path = os.path.join(gt_dir, img_name + '.txt') new_img_path = os.path.join(new_img_dir, img_name) visu_path = os.path.join(visu_dir, img_name) new_gt_path = os.path.join(new_gt_dir, 'gt_' + img_name.split('.')[0] + '.txt') polygons, words = _read_gt(gt_path) # print(img_name) if angle == 90: (h, w) = img.shape[:2] img = cv2.transpose(img) img = cv2.flip(img,flipCode=0) # M = cv2.getRotationMatrix2D(center, 90, 1) # img = cv2.warpAffine(img, M, (h, w)) new_polygons = [[polygon[1], w-polygon[0], polygon[3], w-polygon[2], polygon[5], w-polygon[4], polygon[7], w-polygon[6]] for polygon in polygons] else: img, new_polygons = _rotate_image(img, polygons, angle) format_new_gt(new_polygons, words, new_gt_path) # visu_gt(img, new_polygons, visu_path) cv2.imwrite(new_img_path, img)
6,623
36.213483
200
py
MaskTextSpotterV3
MaskTextSpotterV3-master/tools/test_net.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. # Set up custom environment before nearly anything else is imported # NOTE: this should be the first import (no not reorder) from maskrcnn_benchmark.utils.env import setup_environment # noqa F401 isort:skip import argparse import os import torch from maskrcnn_benchmark.config import cfg from maskrcnn_benchmark.data import make_data_loader from maskrcnn_benchmark.engine.text_inference import inference from maskrcnn_benchmark.modeling.detector import build_detection_model from maskrcnn_benchmark.utils.checkpoint import DetectronCheckpointer from maskrcnn_benchmark.utils.collect_env import collect_env_info from maskrcnn_benchmark.utils.comm import synchronize, get_rank from maskrcnn_benchmark.utils.logging import setup_logger from maskrcnn_benchmark.utils.miscellaneous import mkdir # Check if we can enable mixed-precision via apex.amp try: from apex import amp except ImportError: raise ImportError('Use APEX for mixed precision via apex.amp') def main(): parser = argparse.ArgumentParser(description="PyTorch Object Detection Inference") parser.add_argument( "--config-file", default="./configs/seq.yaml", metavar="FILE", help="path to config file", ) parser.add_argument("--local_rank", type=int, default=0) parser.add_argument( "opts", help="Modify config options using the command-line", default=None, nargs=argparse.REMAINDER, ) args = parser.parse_args() num_gpus = int(os.environ["WORLD_SIZE"]) if "WORLD_SIZE" in os.environ else 1 distributed = num_gpus > 1 if distributed: torch.cuda.set_device(args.local_rank) torch.distributed.deprecated.init_process_group( backend="nccl", init_method="env://" ) cfg.merge_from_file(args.config_file) cfg.merge_from_list(args.opts) cfg.freeze() save_dir = "" logger = setup_logger("maskrcnn_benchmark", save_dir, get_rank()) logger.info("Using {} GPUs".format(num_gpus)) logger.info(cfg) logger.info("Collecting env info (might take some time)") logger.info("\n" + collect_env_info()) model = build_detection_model(cfg) model.to(cfg.MODEL.DEVICE) # Initialize mixed-precision if necessary use_mixed_precision = cfg.DTYPE == 'float16' amp_handle = amp.init(enabled=use_mixed_precision, verbose=cfg.AMP_VERBOSE) checkpointer = DetectronCheckpointer(cfg, model) _ = checkpointer.load(cfg.MODEL.WEIGHT) iou_types = ("bbox",) if cfg.MODEL.MASK_ON: iou_types = iou_types + ("segm",) output_folders = [None] * len(cfg.DATASETS.TEST) if cfg.OUTPUT_DIR: dataset_names = cfg.DATASETS.TEST for idx, dataset_name in enumerate(dataset_names): output_folder = os.path.join(cfg.OUTPUT_DIR, "inference", dataset_name) mkdir(output_folder) output_folders[idx] = output_folder data_loaders_val = make_data_loader(cfg, is_train=False, is_distributed=distributed) model_name = cfg.MODEL.WEIGHT.split('/')[-1] for output_folder, data_loader_val in zip(output_folders, data_loaders_val): inference( model, data_loader_val, iou_types=iou_types, box_only=cfg.MODEL.RPN_ONLY, device=cfg.MODEL.DEVICE, expected_results=cfg.TEST.EXPECTED_RESULTS, expected_results_sigma_tol=cfg.TEST.EXPECTED_RESULTS_SIGMA_TOL, output_folder=output_folder, model_name=model_name, cfg=cfg, ) synchronize() if __name__ == "__main__": main()
3,686
34.451923
88
py
MaskTextSpotterV3
MaskTextSpotterV3-master/tools/demo.py
import os import cv2 import torch from torchvision import transforms as T from maskrcnn_benchmark.modeling.detector import build_detection_model from maskrcnn_benchmark.utils.checkpoint import DetectronCheckpointer from maskrcnn_benchmark.structures.image_list import to_image_list from maskrcnn_benchmark.config import cfg from maskrcnn_benchmark.utils.chars import getstr_grid, get_tight_rect from PIL import Image import numpy as np import argparse class TextDemo(object): def __init__( self, cfg, confidence_threshold=0.7, min_image_size=224, output_polygon=True ): self.cfg = cfg.clone() self.model = build_detection_model(cfg) self.model.eval() self.device = torch.device(cfg.MODEL.DEVICE) self.model.to(self.device) self.min_image_size = min_image_size checkpointer = DetectronCheckpointer(cfg, self.model) _ = checkpointer.load(cfg.MODEL.WEIGHT) self.transforms = self.build_transform() self.cpu_device = torch.device("cpu") self.confidence_threshold = confidence_threshold self.output_polygon = output_polygon def build_transform(self): """ Creates a basic transformation that was used to train the models """ cfg = self.cfg # we are loading images with OpenCV, so we don't need to convert them # to BGR, they are already! So all we need to do is to normalize # by 255 if we want to convert to BGR255 format, or flip the channels # if we want it to be in RGB in [0-1] range. if cfg.INPUT.TO_BGR255: to_bgr_transform = T.Lambda(lambda x: x * 255) else: to_bgr_transform = T.Lambda(lambda x: x[[2, 1, 0]]) normalize_transform = T.Normalize( mean=cfg.INPUT.PIXEL_MEAN, std=cfg.INPUT.PIXEL_STD ) transform = T.Compose( [ T.ToPILImage(), T.Resize(self.min_image_size), T.ToTensor(), to_bgr_transform, normalize_transform, ] ) return transform def run_on_opencv_image(self, image): """ Arguments: image (np.ndarray): an image as returned by OpenCV Returns: result_polygons (list): detection results result_words (list): recognition results """ result_polygons, result_words = self.compute_prediction(image) return result_polygons, result_words def compute_prediction(self, original_image): # apply pre-processing to image image = self.transforms(original_image) # convert to an ImageList, padded so that it is divisible by # cfg.DATALOADER.SIZE_DIVISIBILITY image_list = to_image_list(image, self.cfg.DATALOADER.SIZE_DIVISIBILITY) image_list = image_list.to(self.device) # compute predictions with torch.no_grad(): predictions, _, _ = self.model(image_list) global_predictions = predictions[0] char_predictions = predictions[1] char_mask = char_predictions['char_mask'] char_boxes = char_predictions['boxes'] words, rec_scores = self.process_char_mask(char_mask, char_boxes) seq_words = char_predictions['seq_outputs'] seq_scores = char_predictions['seq_scores'] global_predictions = [o.to(self.cpu_device) for o in global_predictions] # always single image is passed at a time global_prediction = global_predictions[0] # reshape prediction (a BoxList) into the original image size height, width = original_image.shape[:-1] global_prediction = global_prediction.resize((width, height)) boxes = global_prediction.bbox.tolist() scores = global_prediction.get_field("scores").tolist() masks = global_prediction.get_field("mask").cpu().numpy() result_polygons = [] result_words = [] for k, box in enumerate(boxes): score = scores[k] if score < self.confidence_threshold: continue box = list(map(int, box)) mask = masks[k,0,:,:] polygon = self.mask2polygon(mask, box, original_image.shape, threshold=0.5, output_polygon=self.output_polygon) if polygon is None: polygon = [box[0], box[1], box[2], box[1], box[2], box[3], box[0], box[3]] result_polygons.append(polygon) word = words[k] rec_score = rec_scores[k] seq_word = seq_words[k] seq_char_scores = seq_scores[k] seq_score = sum(seq_char_scores) / float(len(seq_char_scores)) if seq_score > rec_score: result_words.append(seq_word) else: result_words.append(word) return result_polygons, result_words def process_char_mask(self, char_masks, boxes, threshold=192): texts, rec_scores = [], [] for index in range(char_masks.shape[0]): box = list(boxes[index]) box = list(map(int, box)) text, rec_score, _, _ = getstr_grid(char_masks[index,:,:,:].copy(), box, threshold=threshold) texts.append(text) rec_scores.append(rec_score) return texts, rec_scores def mask2polygon(self, mask, box, im_size, threshold=0.5, output_polygon=True): # mask 32*128 image_width, image_height = im_size[1], im_size[0] box_h = box[3] - box[1] box_w = box[2] - box[0] cls_polys = (mask*255).astype(np.uint8) poly_map = np.array(Image.fromarray(cls_polys).resize((box_w, box_h))) poly_map = poly_map.astype(np.float32) / 255 poly_map=cv2.GaussianBlur(poly_map,(3,3),sigmaX=3) ret, poly_map = cv2.threshold(poly_map,0.5,1,cv2.THRESH_BINARY) if output_polygon: SE1=cv2.getStructuringElement(cv2.MORPH_RECT,(3,3)) poly_map = cv2.erode(poly_map,SE1) poly_map = cv2.dilate(poly_map,SE1); poly_map = cv2.morphologyEx(poly_map,cv2.MORPH_CLOSE,SE1) try: _, contours, _ = cv2.findContours((poly_map * 255).astype(np.uint8), cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE) except: contours, _ = cv2.findContours((poly_map * 255).astype(np.uint8), cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE) if len(contours)==0: print(contours) print(len(contours)) return None max_area=0 max_cnt = contours[0] for cnt in contours: area=cv2.contourArea(cnt) if area > max_area: max_area = area max_cnt = cnt perimeter = cv2.arcLength(max_cnt,True) epsilon = 0.01*cv2.arcLength(max_cnt,True) approx = cv2.approxPolyDP(max_cnt,epsilon,True) pts = approx.reshape((-1,2)) pts[:,0] = pts[:,0] + box[0] pts[:,1] = pts[:,1] + box[1] polygon = list(pts.reshape((-1,))) polygon = list(map(int, polygon)) if len(polygon)<6: return None else: SE1=cv2.getStructuringElement(cv2.MORPH_RECT,(3,3)) poly_map = cv2.erode(poly_map,SE1) poly_map = cv2.dilate(poly_map,SE1); poly_map = cv2.morphologyEx(poly_map,cv2.MORPH_CLOSE,SE1) idy,idx=np.where(poly_map == 1) xy=np.vstack((idx,idy)) xy=np.transpose(xy) hull = cv2.convexHull(xy, clockwise=True) #reverse order of points. if hull is None: return None hull=hull[::-1] #find minimum area bounding box. rect = cv2.minAreaRect(hull) corners = cv2.boxPoints(rect) corners = np.array(corners, dtype="int") pts = get_tight_rect(corners, box[0], box[1], image_height, image_width, 1) polygon = [x * 1.0 for x in pts] polygon = list(map(int, polygon)) return polygon def visualization(self, image, polygons, words): for polygon, word in zip(polygons, words): pts = np.array(polygon, np.int32) pts = pts.reshape((-1,1,2)) xmin = min(pts[:,0,0]) ymin = min(pts[:,0,1]) cv2.polylines(image,[pts],True,(0,0,255)) cv2.putText(image, word, (xmin, ymin), cv2.FONT_HERSHEY_COMPLEX, 1, (0,0,255), 2) def main(args): # update the config options with the config file cfg.merge_from_file(args.config_file) # manual override some options # cfg.merge_from_list(["MODEL.DEVICE", "cpu"]) text_demo = TextDemo( cfg, min_image_size=800, confidence_threshold=0.7, output_polygon=True ) # load image and then run prediction image = cv2.imread(args.image_path) result_polygons, result_words = text_demo.run_on_opencv_image(image) text_demo.visualization(image, result_polygons, result_words) cv2.imwrite(args.visu_path, image) if __name__ == "__main__": parser = argparse.ArgumentParser(description='parameters for demo') parser.add_argument("--config-file", type=str, default='configs/mixtrain/seg_rec_poly_fuse_feature.yaml') parser.add_argument("--image_path", type=str, default='./demo_images/demo.jpg') parser.add_argument("--visu_path", type=str, default='./demo_images/demo_results.jpg') args = parser.parse_args() main(args)
9,628
39.288703
123
py
MaskTextSpotterV3
MaskTextSpotterV3-master/tools/train_net.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. r""" Basic training script for PyTorch """ # Set up custom environment before nearly anything else is imported # NOTE: this should be the first import (no not reorder) from maskrcnn_benchmark.utils.env import setup_environment # noqa F401 isort:skip import argparse import os import torch from maskrcnn_benchmark.config import cfg from maskrcnn_benchmark.data import make_data_loader from maskrcnn_benchmark.solver import make_lr_scheduler from maskrcnn_benchmark.solver import make_optimizer from maskrcnn_benchmark.engine.trainer import do_train from maskrcnn_benchmark.modeling.detector import build_detection_model from maskrcnn_benchmark.utils.checkpoint import DetectronCheckpointer from maskrcnn_benchmark.utils.collect_env import collect_env_info from maskrcnn_benchmark.utils.comm import synchronize, get_rank from maskrcnn_benchmark.utils.imports import import_file from maskrcnn_benchmark.utils.logging import setup_logger, Logger from maskrcnn_benchmark.utils.miscellaneous import mkdir # See if we can use apex.DistributedDataParallel instead of the torch default, # and enable mixed-precision via apex.amp try: from apex import amp except ImportError: raise ImportError('Use APEX for multi-precision via apex.amp') def train(cfg, local_rank, distributed): model = build_detection_model(cfg) device = torch.device(cfg.MODEL.DEVICE) model.to(device) optimizer = make_optimizer(cfg, model) scheduler = make_lr_scheduler(cfg, optimizer) # Initialize mixed-precision training use_mixed_precision = cfg.DTYPE == "float16" amp_opt_level = 'O1' if use_mixed_precision else 'O0' model, optimizer = amp.initialize(model, optimizer, opt_level=amp_opt_level) if distributed: model = torch.nn.parallel.DistributedDataParallel( model, device_ids=[local_rank], output_device=local_rank, # this should be removed if we update BatchNorm stats broadcast_buffers=False, # find_unused_parameters=True ) arguments = {} arguments["iteration"] = 0 output_dir = cfg.OUTPUT_DIR save_to_disk = get_rank() == 0 checkpointer = DetectronCheckpointer( cfg, model, optimizer, scheduler, output_dir, save_to_disk ) extra_checkpoint_data = checkpointer.load(cfg.MODEL.WEIGHT, resume=cfg.SOLVER.RESUME) if cfg.SOLVER.RESUME: arguments.update(extra_checkpoint_data) data_loader = make_data_loader( cfg, is_train=True, is_distributed=distributed, start_iter=arguments["iteration"], ) checkpoint_period = cfg.SOLVER.CHECKPOINT_PERIOD tb_logger = Logger(cfg.OUTPUT_DIR, local_rank) do_train( model, data_loader, optimizer, scheduler, checkpointer, device, checkpoint_period, arguments, tb_logger, cfg, local_rank, ) return model def main(): parser = argparse.ArgumentParser(description="PyTorch Object Detection Training") parser.add_argument( "--config-file", default="", metavar="FILE", help="path to config file", type=str, ) parser.add_argument("--local_rank", type=int, default=0) parser.add_argument( "--skip-test", dest="skip_test", help="Do not test the final model", action="store_true", ) parser.add_argument( "opts", help="Modify config options using the command-line", default=None, nargs=argparse.REMAINDER, ) args = parser.parse_args() num_gpus = int(os.environ["WORLD_SIZE"]) if "WORLD_SIZE" in os.environ else 1 args.distributed = num_gpus > 1 if args.distributed: torch.cuda.set_device(args.local_rank) torch.distributed.init_process_group( backend="nccl", init_method="env://" ) cfg.merge_from_file(args.config_file) cfg.merge_from_list(args.opts) cfg.freeze() output_dir = cfg.OUTPUT_DIR if output_dir: mkdir(output_dir) local_rank = get_rank() logger = setup_logger("maskrcnn_benchmark", output_dir, local_rank) if local_rank == 0: logger.info("Using {} GPUs".format(num_gpus)) logger.info(args) logger.info("Collecting env info (might take some time)") logger.info("\n" + collect_env_info()) logger.info("Loaded configuration file {}".format(args.config_file)) with open(args.config_file, "r") as cf: config_str = "\n" + cf.read() logger.info(config_str) logger.info("Running with config:\n{}".format(cfg)) model = train(cfg, args.local_rank, args.distributed) if __name__ == "__main__": main()
4,818
30.292208
89
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/solver/lr_scheduler.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. from bisect import bisect_right import torch # FIXME ideally this would be achieved with a CombinedLRScheduler, # separating MultiStepLR with WarmupLR # but the current LRScheduler design doesn't allow it class WarmupMultiStepLR(torch.optim.lr_scheduler._LRScheduler): def __init__( self, optimizer, milestones, gamma=0.1, warmup_factor=1.0 / 3, warmup_iters=500, warmup_method="linear", last_epoch=-1, pow_schedule_mode = False, max_iter = 300000, lr_pow = 0.9 ): if not list(milestones) == sorted(milestones): raise ValueError( "Milestones should be a list of" " increasing integers. Got {}", milestones, ) if warmup_method not in ("constant", "linear"): raise ValueError( "Only 'constant' or 'linear' warmup_method accepted" "got {}".format(warmup_method) ) self.milestones = milestones self.gamma = gamma self.warmup_factor = warmup_factor self.warmup_iters = warmup_iters self.warmup_method = warmup_method self.pow_schedule_mode = pow_schedule_mode self.max_iter = max_iter self.lr_pow = lr_pow super(WarmupMultiStepLR, self).__init__(optimizer, last_epoch) def get_lr(self): warmup_factor = 1 if self.last_epoch < self.warmup_iters: if self.warmup_method == "constant": warmup_factor = self.warmup_factor elif self.warmup_method == "linear": alpha = self.last_epoch / self.warmup_iters warmup_factor = self.warmup_factor * (1 - alpha) + alpha if self.pow_schedule_mode: scale_running_lr = ((1. - float(self.last_epoch) / self.max_iter) ** self.lr_pow) return [ base_lr * warmup_factor * scale_running_lr for base_lr in self.base_lrs ] else: return [ base_lr * warmup_factor * self.gamma ** bisect_right(self.milestones, self.last_epoch) for base_lr in self.base_lrs ]
2,292
33.742424
93
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/solver/__init__.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. from .build import make_optimizer from .build import make_lr_scheduler from .lr_scheduler import WarmupMultiStepLR
187
36.6
71
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/solver/build.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import torch from .lr_scheduler import WarmupMultiStepLR def make_optimizer(cfg, model): params = [] for key, value in model.named_parameters(): if not value.requires_grad: continue lr = cfg.SOLVER.BASE_LR weight_decay = cfg.SOLVER.WEIGHT_DECAY if "bias" in key: lr = cfg.SOLVER.BASE_LR * cfg.SOLVER.BIAS_LR_FACTOR weight_decay = cfg.SOLVER.WEIGHT_DECAY_BIAS params += [{"params": [value], "lr": lr, "weight_decay": weight_decay}] if cfg.SOLVER.USE_ADAM: optimizer = torch.optim.Adam(model.parameters(), lr=lr) else: optimizer = torch.optim.SGD(params, lr, momentum=cfg.SOLVER.MOMENTUM) return optimizer def make_lr_scheduler(cfg, optimizer): return WarmupMultiStepLR( optimizer, cfg.SOLVER.STEPS, cfg.SOLVER.GAMMA, warmup_factor=cfg.SOLVER.WARMUP_FACTOR, warmup_iters=cfg.SOLVER.WARMUP_ITERS, warmup_method=cfg.SOLVER.WARMUP_METHOD, pow_schedule_mode = cfg.SOLVER.POW_SCHEDULE, max_iter = cfg.SOLVER.MAX_ITER, )
1,176
29.973684
79
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/config/defaults.py
#!/usr/bin/env python3 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import os from yacs.config import CfgNode as CN # ----------------------------------------------------------------------------- # Convention about Training / Test specific parameters # ----------------------------------------------------------------------------- # Whenever an argument can be either used for training or for testing, the # corresponding name will be post-fixed by a _TRAIN for a training parameter, # or _TEST for a test-specific parameter. # For example, the number of images during training will be # IMAGES_PER_BATCH_TRAIN, while the number of images for testing will be # IMAGES_PER_BATCH_TEST # ----------------------------------------------------------------------------- # Config definition # ----------------------------------------------------------------------------- _C = CN() _C.MODEL = CN() _C.MODEL.RPN_ONLY = False _C.MODEL.MASK_ON = False _C.MODEL.SEG_ON = False _C.MODEL.CHAR_MASK_ON = False _C.MODEL.DEVICE = "cuda" _C.MODEL.META_ARCHITECTURE = "GeneralizedRCNN" _C.MODEL.TRAIN_DETECTION_ONLY = False _C.MODEL.RESNET34 = False # If the WEIGHT starts with a catalog://, like :R-50, the code will look for # the path in paths_catalog. Else, it will use it as the specified absolute # path _C.MODEL.WEIGHT = "" _C.SEQUENCE = CN() _C.SEQUENCE.SEQ_ON = False _C.SEQUENCE.NUM_CHAR = 38 _C.SEQUENCE.BOS_TOKEN = 0 _C.SEQUENCE.MAX_LENGTH = 32 _C.SEQUENCE.TEACHER_FORCE_RATIO = 1.0 _C.SEQUENCE.TWO_CONV = False _C.SEQUENCE.MEAN_SCORE = False _C.SEQUENCE.RESIZE_HEIGHT = 16 _C.SEQUENCE.RESIZE_WIDTH = 64 # ----------------------------------------------------------------------------- # INPUT # ----------------------------------------------------------------------------- _C.INPUT = CN() # Size of the smallest side of the image during training _C.INPUT.MIN_SIZE_TRAIN = (800,) # (800,) # Maximum size of the side of the image during training _C.INPUT.MAX_SIZE_TRAIN = 1333 # Size of the smallest side of the image during testing _C.INPUT.MIN_SIZE_TEST = 800 # Maximum size of the side of the image during testing _C.INPUT.MAX_SIZE_TEST = 1333 # Values to be used for image normalization _C.INPUT.PIXEL_MEAN = [102.9801, 115.9465, 122.7717] # Values to be used for image normalization _C.INPUT.PIXEL_STD = [1.0, 1.0, 1.0] # Convert image to BGR format (for Caffe2 models), in range 0-255 _C.INPUT.TO_BGR255 = True _C.INPUT.STRICT_RESIZE = False # ----------------------------------------------------------------------------- # Dataset # ----------------------------------------------------------------------------- _C.DATASETS = CN() # List of the dataset names for training, as present in paths_catalog.py _C.DATASETS.TRAIN = () # List of the dataset names for testing, as present in paths_catalog.py _C.DATASETS.TEST = () _C.DATASETS.RATIOS = [] _C.DATASETS.AUG = False _C.DATASETS.RANDOM_CROP_PROB = 0.0 _C.DATASETS.IGNORE_DIFFICULT = False _C.DATASETS.FIX_CROP = False _C.DATASETS.CROP_SIZE = (512, 512) _C.DATASETS.MAX_ROTATE_THETA = 30 _C.DATASETS.FIX_ROTATE = False # ----------------------------------------------------------------------------- # DataLoader # ----------------------------------------------------------------------------- _C.DATALOADER = CN() # Number of data loading threads _C.DATALOADER.NUM_WORKERS = 4 # If > 0, this enforces that each collated batch should have a size divisible # by SIZE_DIVISIBILITY _C.DATALOADER.SIZE_DIVISIBILITY = 0 # If True, each batch should contain only images for which the aspect ratio # is compatible. This groups portrait images together, and landscape images # are not batched with portrait images. _C.DATALOADER.ASPECT_RATIO_GROUPING = True # ---------------------------------------------------------------------------- # # Backbone options # ---------------------------------------------------------------------------- # _C.MODEL.BACKBONE = CN() # The backbone conv body to use # The string must match a function that is imported in modeling.model_builder # (e.g., 'FPN.add_fpn_ResNet101_conv5_body' to specify a ResNet-101-FPN # backbone) _C.MODEL.BACKBONE.CONV_BODY = "R-50-C4" # Add StopGrad at a specified stage so the bottom layers are frozen _C.MODEL.BACKBONE.FREEZE_CONV_BODY_AT = 2 _C.MODEL.BACKBONE.OUT_CHANNELS = 256 * 4 # ---------------------------------------------------------------------------- # # ResNe[X]t options (ResNets = {ResNet, ResNeXt} # Note that parts of a resnet may be used for both the backbone and the head # These options apply to both # ---------------------------------------------------------------------------- # _C.MODEL.RESNETS = CN() # Number of groups to use; 1 ==> ResNet; > 1 ==> ResNeXt _C.MODEL.RESNETS.NUM_GROUPS = 1 # Baseline width of each group _C.MODEL.RESNETS.WIDTH_PER_GROUP = 64 # Place the stride 2 conv on the 1x1 filter # Use True only for the original MSRA ResNet; use False for C2 and Torch models _C.MODEL.RESNETS.STRIDE_IN_1X1 = True # Residual transformation function _C.MODEL.RESNETS.TRANS_FUNC = "BottleneckWithFixedBatchNorm" # ResNet's stem function (conv1 and pool1) _C.MODEL.RESNETS.STEM_FUNC = "StemWithFixedBatchNorm" # Apply dilation in stage "res5" _C.MODEL.RESNETS.RES5_DILATION = 1 _C.MODEL.RESNETS.BACKBONE_OUT_CHANNELS = 256 * 4 _C.MODEL.RESNETS.RES2_OUT_CHANNELS = 256 _C.MODEL.RESNETS.STEM_OUT_CHANNELS = 64 _C.MODEL.RESNETS.STAGE_WITH_DCN = (False, False, False, False) _C.MODEL.RESNETS.WITH_MODULATED_DCN = False _C.MODEL.RESNETS.DEFORMABLE_GROUPS = 1 _C.MODEL.RESNETS.LAYERS = (3, 4, 6, 3) # ---------------------------------------------------------------------------- # # FPN options # ---------------------------------------------------------------------------- # _C.MODEL.FPN = CN() _C.MODEL.FPN.USE_GN = False _C.MODEL.FPN.USE_RELU = False # ---------------------------------------------------------------------------- # # RPN options # ---------------------------------------------------------------------------- # _C.MODEL.RPN = CN() _C.MODEL.RPN.USE_FPN = False # Base RPN anchor sizes given in absolute pixels w.r.t. the scaled network input _C.MODEL.RPN.ANCHOR_SIZES = (32, 64, 128, 256, 512) # Stride of the feature map that RPN is attached. # For FPN, number of strides should match number of scales _C.MODEL.RPN.ANCHOR_STRIDE = (16,) # RPN anchor aspect ratios _C.MODEL.RPN.ASPECT_RATIOS = (0.5, 1.0, 2.0) # Remove RPN anchors that go outside the image by RPN_STRADDLE_THRESH pixels # Set to -1 or a large value, e.g. 100000, to disable pruning anchors _C.MODEL.RPN.STRADDLE_THRESH = 0 # Minimum overlap required between an anchor and ground-truth box for the # (anchor, gt box) pair to be a positive example (IoU >= FG_IOU_THRESHOLD # ==> positive RPN example) _C.MODEL.RPN.FG_IOU_THRESHOLD = 0.7 # Maximum overlap allowed between an anchor and ground-truth box for the # (anchor, gt box) pair to be a negative examples (IoU < BG_IOU_THRESHOLD # ==> negative RPN example) _C.MODEL.RPN.BG_IOU_THRESHOLD = 0.3 # Total number of RPN examples per image _C.MODEL.RPN.BATCH_SIZE_PER_IMAGE = 256 # Target fraction of foreground (positive) examples per RPN minibatch _C.MODEL.RPN.POSITIVE_FRACTION = 0.5 # Number of top scoring RPN proposals to keep before applying NMS # When FPN is used, this is *per FPN level* (not total) _C.MODEL.RPN.PRE_NMS_TOP_N_TRAIN = 12000 _C.MODEL.RPN.PRE_NMS_TOP_N_TEST = 6000 # Number of top scoring RPN proposals to keep after applying NMS _C.MODEL.RPN.POST_NMS_TOP_N_TRAIN = 2000 _C.MODEL.RPN.POST_NMS_TOP_N_TEST = 1000 # NMS threshold used on RPN proposals _C.MODEL.RPN.NMS_THRESH = 0.7 # Proposal height and width both need to be greater than RPN_MIN_SIZE # (a the scale used during training or inference) _C.MODEL.RPN.MIN_SIZE = 0 # Number of top scoring RPN proposals to keep after combining proposals from # all FPN levels _C.MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN = 2000 _C.MODEL.RPN.FPN_POST_NMS_TOP_N_TEST = 2000 _C.MODEL.SEG = CN() _C.MODEL.SEG.USE_FPN = False _C.MODEL.SEG.USE_FUSE_FEATURE = False # Total number of SEG examples per image _C.MODEL.SEG.BATCH_SIZE_PER_IMAGE = 256 # Target fraction of foreground (positive) examples per SEG minibatch _C.MODEL.SEG.POSITIVE_FRACTION = 0.5 # NMS threshold used on SEG proposals _C.MODEL.SEG.BINARY_THRESH = 0.5 _C.MODEL.SEG.USE_MULTIPLE_THRESH = False _C.MODEL.SEG.MULTIPLE_THRESH = (0.2, 0.3, 0.5, 0.7) _C.MODEL.SEG.BOX_THRESH = 0.7 # Proposal height and width both need to be greater than RPN_MIN_SIZE # (a the scale used during training or inference) _C.MODEL.SEG.MIN_SIZE = 0 _C.MODEL.SEG.SHRINK_RATIO = 0.5 # Number of top scoring RPN proposals to keep after combining proposals from # all FPN levels _C.MODEL.SEG.TOP_N_TRAIN = 1000 _C.MODEL.SEG.TOP_N_TEST = 1000 _C.MODEL.SEG.AUG_PROPOSALS = False _C.MODEL.SEG.IGNORE_DIFFICULT = True _C.MODEL.SEG.EXPAND_RATIO = 1.6 _C.MODEL.SEG.BOX_EXPAND_RATIO = 1.5 _C.MODEL.SEG.USE_SEG_POLY = False _C.MODEL.SEG.USE_PPM = False # ---------------------------------------------------------------------------- # # ROI HEADS options # ---------------------------------------------------------------------------- # _C.MODEL.ROI_HEADS = CN() _C.MODEL.ROI_HEADS.USE_FPN = False # Overlap threshold for an RoI to be considered foreground (if >= FG_IOU_THRESHOLD) _C.MODEL.ROI_HEADS.FG_IOU_THRESHOLD = 0.5 # Overlap threshold for an RoI to be considered background # (class = 0 if overlap in [0, BG_IOU_THRESHOLD)) _C.MODEL.ROI_HEADS.BG_IOU_THRESHOLD = 0.5 # Default weights on (dx, dy, dw, dh) for normalizing bbox regression targets # These are empirically chosen to approximately lead to unit variance targets _C.MODEL.ROI_HEADS.BBOX_REG_WEIGHTS = (10.0, 10.0, 5.0, 5.0) # RoI minibatch size *per image* (number of regions of interest [ROIs]) # Total number of RoIs per training minibatch = # TRAIN.BATCH_SIZE_PER_IM * TRAIN.IMS_PER_BATCH * NUM_GPUS # E.g., a common configuration is: 512 * 2 * 8 = 8192 _C.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 512 # Target fraction of RoI minibatch that is labeled foreground (i.e. class > 0) _C.MODEL.ROI_HEADS.POSITIVE_FRACTION = 0.25 # Only used on test mode # Minimum score threshold (assuming scores in a [0, 1] range); a value chosen to # balance obtaining high recall with not having too many low precision # detections that will slow down inference post processing steps (like NMS) # _C.MODEL.ROI_HEADS.SCORE_THRESH = 0.05 _C.MODEL.ROI_HEADS.SCORE_THRESH = 0.0 # Overlap threshold used for non-maximum suppression (suppress boxes with # IoU >= this threshold) _C.MODEL.ROI_HEADS.NMS = 0.5 # Maximum number of detections to return per image (100 is based on the limit # established for the COCO dataset) _C.MODEL.ROI_HEADS.DETECTIONS_PER_IMG = 100 _C.MODEL.ROI_BOX_HEAD = CN() _C.MODEL.ROI_BOX_HEAD.FEATURE_EXTRACTOR = "ResNet50Conv5ROIFeatureExtractor" _C.MODEL.ROI_BOX_HEAD.PREDICTOR = "FastRCNNPredictor" _C.MODEL.ROI_BOX_HEAD.POOLER_RESOLUTION = 14 _C.MODEL.ROI_BOX_HEAD.POOLER_SAMPLING_RATIO = 0 _C.MODEL.ROI_BOX_HEAD.POOLER_SCALES = (1.0 / 16,) _C.MODEL.ROI_BOX_HEAD.NUM_CLASSES = 81 # Hidden layer dimension when using an MLP for the RoI box head _C.MODEL.ROI_BOX_HEAD.MLP_HEAD_DIM = 1024 _C.MODEL.ROI_BOX_HEAD.USE_REGRESSION = True _C.MODEL.ROI_BOX_HEAD.INFERENCE_USE_BOX = True _C.MODEL.ROI_BOX_HEAD.USE_MASKED_FEATURE = False _C.MODEL.ROI_BOX_HEAD.SOFT_MASKED_FEATURE_RATIO = 0. _C.MODEL.ROI_BOX_HEAD.MIX_OPTION = "" _C.MODEL.ROI_MASK_HEAD = CN() _C.MODEL.ROI_MASK_HEAD.FEATURE_EXTRACTOR = "ResNet50Conv5ROIFeatureExtractor" _C.MODEL.ROI_MASK_HEAD.PREDICTOR = "MaskRCNNC4Predictor" _C.MODEL.ROI_MASK_HEAD.POOLER_RESOLUTION = 14 _C.MODEL.ROI_MASK_HEAD.POOLER_RESOLUTION_H = 32 _C.MODEL.ROI_MASK_HEAD.POOLER_RESOLUTION_W = 128 _C.MODEL.ROI_MASK_HEAD.POOLER_SAMPLING_RATIO = 0 _C.MODEL.ROI_MASK_HEAD.POOLER_SCALES = (1.0 / 16,) _C.MODEL.ROI_MASK_HEAD.MLP_HEAD_DIM = 1024 _C.MODEL.ROI_MASK_HEAD.CONV_LAYERS = (256, 256, 256, 256) _C.MODEL.ROI_MASK_HEAD.RESOLUTION = 14 _C.MODEL.ROI_MASK_HEAD.RESOLUTION_H = 32 _C.MODEL.ROI_MASK_HEAD.RESOLUTION_W = 128 _C.MODEL.ROI_MASK_HEAD.SHARE_BOX_FEATURE_EXTRACTOR = True _C.MODEL.ROI_MASK_HEAD.CHAR_NUM_CLASSES = 38 _C.MODEL.ROI_MASK_HEAD.USE_WEIGHTED_CHAR_MASK = False _C.MODEL.ROI_MASK_HEAD.MASK_BATCH_SIZE_PER_IM = 64 _C.MODEL.ROI_MASK_HEAD.USE_MASKED_FEATURE = False _C.MODEL.ROI_MASK_HEAD.SOFT_MASKED_FEATURE_RATIO = 0. _C.MODEL.ROI_MASK_HEAD.MIX_OPTION = "" # ---------------------------------------------------------------------------- # # Solver # ---------------------------------------------------------------------------- # _C.SOLVER = CN() _C.SOLVER.MAX_ITER = 40000 _C.SOLVER.BASE_LR = 0.001 _C.SOLVER.BIAS_LR_FACTOR = 2 _C.SOLVER.MOMENTUM = 0.9 _C.SOLVER.WEIGHT_DECAY = 0.0005 _C.SOLVER.WEIGHT_DECAY_BIAS = 0 _C.SOLVER.GAMMA = 0.1 _C.SOLVER.STEPS = (30000,) _C.SOLVER.WARMUP_FACTOR = 1.0 / 3 _C.SOLVER.WARMUP_ITERS = 500 _C.SOLVER.WARMUP_METHOD = "linear" _C.SOLVER.CHECKPOINT_PERIOD = 5000 # Number of images per batch # This is global, so if we have 8 GPUs and IMS_PER_BATCH = 16, each GPU will # see 2 images per batch _C.SOLVER.IMS_PER_BATCH = 16 _C.SOLVER.RESUME = True _C.SOLVER.USE_ADAM = False _C.SOLVER.POW_SCHEDULE = False _C.SOLVER.DISPLAY_FREQ = 20 # ---------------------------------------------------------------------------- # # Specific test options # ---------------------------------------------------------------------------- # _C.TEST = CN() _C.TEST.EXPECTED_RESULTS = [] _C.TEST.EXPECTED_RESULTS_SIGMA_TOL = 4 # Number of images per batch # This is global, so if we have 8 GPUs and IMS_PER_BATCH = 16, each GPU will # see 2 images per batch _C.TEST.IMS_PER_BATCH = 8 _C.TEST.VIS = False # from 0 to 255 _C.TEST.CHAR_THRESH = 128 # ---------------------------------------------------------------------------- # # Misc options # ---------------------------------------------------------------------------- # _C.OUTPUT_DIR = "." _C.PATHS_CATALOG = os.path.join(os.path.dirname(__file__), "paths_catalog.py") # ---------------------------------------------------------------------------- # # Precision options # ---------------------------------------------------------------------------- # # Precision of input, allowable: (float32, float16) _C.DTYPE = "float32" # Enable verbosity in apex.amp _C.AMP_VERBOSE = False
14,283
37.294906
83
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/config/__init__.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. from .defaults import _C as cfg
104
34
71
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/config/paths_catalog.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. """Centralized catalog of paths.""" import os class DatasetCatalog(object): DATA_DIR = "datasets" # DATA_DIR = "/share/mhliao/MaskTextSpotterV3/datasets/" DATASETS = { "coco_2014_train": ( "coco/train2014", "coco/annotations/instances_train2014.json", ), "coco_2014_val": ("coco/val2014", "coco/annotations/instances_val2014.json"), "coco_2014_minival": ( "coco/val2014", "coco/annotations/instances_minival2014.json", ), "coco_2014_valminusminival": ( "coco/val2014", "coco/annotations/instances_valminusminival2014.json", ), "icdar_2013_train": ("icdar2013/train_images", "icdar2013/train_gts"), "icdar_2013_test": ("icdar2013/test_images", "icdar2013/test_gts"), "rotated_ic13_test_0": ("icdar2013/rotated_test_images_0", "icdar2013/rotated_test_gts_0"), "rotated_ic13_test_15": ("icdar2013/rotated_test_images_15", "icdar2013/rotated_test_gts_15"), "rotated_ic13_test_30": ("icdar2013/rotated_test_images_30", "icdar2013/rotated_test_gts_30"), "rotated_ic13_test_45": ("icdar2013/rotated_test_images_45", "icdar2013/rotated_test_gts_45"), "rotated_ic13_test_60": ("icdar2013/rotated_test_images_60", "icdar2013/rotated_test_gts_60"), "rotated_ic13_test_75": ("icdar2013/rotated_test_images_75", "icdar2013/rotated_test_gts_75"), "rotated_ic13_test_85": ("icdar2013/rotated_test_images_85", "icdar2013/rotated_test_gts_85"), "rotated_ic13_test_90": ("icdar2013/rotated_test_images_90", "icdar2013/rotated_test_gts_90"), "rotated_ic13_test_-15": ("icdar2013/rotated_test_images_-15", "icdar2013/rotated_test_gts_-15"), "rotated_ic13_test_-30": ("icdar2013/rotated_test_images_-30", "icdar2013/rotated_test_gts_-30"), "rotated_ic13_test_-45": ("icdar2013/rotated_test_images_-45", "icdar2013/rotated_test_gts_-45"), "rotated_ic13_test_-60": ("icdar2013/rotated_test_images_-60", "icdar2013/rotated_test_gts_-60"), "rotated_ic13_test_-75": ("icdar2013/rotated_test_images_-75", "icdar2013/rotated_test_gts_-75"), "rotated_ic13_test_-90": ("icdar2013/rotated_test_images_-90", "icdar2013/rotated_test_gts_-90"), "icdar_2015_train": ("icdar2015/train_images", "icdar2015/train_gts"), "icdar_2015_test": ( "icdar2015/test_images", # "icdar2015/test_gts", ), "synthtext_train": ("synthtext/train_images", "synthtext/train_gts"), "synthtext_test": ("synthtext/test_images", "synthtext/test_gts"), "total_text_train": ("total_text/train_images", "total_text/train_gts"), "td500_train": ("TD_TR/TD500/train_images", "TD500/train_gts"), "td500_test": ("TD_TR/TD500/test_images", ), "tr400_train": ("TD_TR/TR400/train_images", "TR400/train_gts"), "total_text_test": ( "total_text/test_images", # "total_text/test_gts", ), "scut-eng-char_train": ( "scut-eng-char/train_images", "scut-eng-char/train_gts", ), } @staticmethod def get(name): if "coco" in name: data_dir = DatasetCatalog.DATA_DIR attrs = DatasetCatalog.DATASETS[name] args = dict( root=os.path.join(data_dir, attrs[0]), ann_file=os.path.join(data_dir, attrs[1]), ) return dict(factory="COCODataset", args=args) elif "icdar_2013" in name: data_dir = DatasetCatalog.DATA_DIR attrs = DatasetCatalog.DATASETS[name] args = dict( use_charann=True, imgs_dir=os.path.join(data_dir, attrs[0]), gts_dir=os.path.join(data_dir, attrs[1]), # imgs_dir='/tmp/icdar2013/icdar2013/train_images', # gts_dir='/tmp/icdar2013/icdar2013/train_gts', ) return dict(args=args, factory="IcdarDataset") elif "rotated_ic13" in name: data_dir = DatasetCatalog.DATA_DIR attrs = DatasetCatalog.DATASETS[name] args = dict( use_charann=True, imgs_dir=os.path.join(data_dir, attrs[0]), gts_dir=os.path.join(data_dir, attrs[1]), ) return dict(args=args, factory="IcdarDataset") elif "icdar_2015" in name: data_dir = DatasetCatalog.DATA_DIR attrs = DatasetCatalog.DATASETS[name] if len(attrs) > 1: gts_dir = os.path.join(data_dir, attrs[1]) else: gts_dir = None args = dict( use_charann=False, imgs_dir=os.path.join(data_dir, attrs[0]), gts_dir=gts_dir, # imgs_dir='/tmp/icdar2015/icdar2015/train_images/', # gts_dir='/tmp/icdar2015/icdar2015/train_gts/', ) return dict(args=args, factory="IcdarDataset") elif "synthtext" in name: data_dir = DatasetCatalog.DATA_DIR attrs = DatasetCatalog.DATASETS[name] args = dict( use_charann=True, list_file_path=os.path.join(data_dir, "synthtext/train_list.txt"), imgs_dir=os.path.join(data_dir, attrs[0]), gts_dir=os.path.join(data_dir, attrs[1]), # imgs_dir='/tmp/synth/SynthText/', # gts_dir='/tmp/synth_gt/SynthText_GT_E2E/', ) return dict(args=args, factory="SynthtextDataset") elif "total_text" in name: data_dir = DatasetCatalog.DATA_DIR # data_dir = '/tmp/total_text/' attrs = DatasetCatalog.DATASETS[name] if len(attrs) > 1: gts_dir = os.path.join(data_dir, attrs[1]) else: gts_dir = None args = dict( use_charann=False, imgs_dir=os.path.join(data_dir, attrs[0]), gts_dir=gts_dir, # imgs_dir='/tmp/total_text/total_text/train_images/', # gts_dir='/tmp/total_text/total_text/train_gts/', ) return dict(args=args, factory="TotaltextDataset") elif "scut-eng-char" in name: data_dir = DatasetCatalog.DATA_DIR attrs = DatasetCatalog.DATASETS[name] args = dict( use_charann=True, imgs_dir=os.path.join(data_dir, attrs[0]), gts_dir=os.path.join(data_dir, attrs[1]), # imgs_dir='/tmp/scut-eng-char/scut-eng-char/train_images/', # gts_dir='/tmp/scut-eng-char/scut-eng-char/train_gts/', ) return dict(args=args, factory="ScutDataset") elif "td500" in name: data_dir = DatasetCatalog.DATA_DIR attrs = DatasetCatalog.DATASETS[name] if len(attrs) > 1: gts_dir = os.path.join(data_dir, attrs[1]) else: gts_dir = None args = dict( use_charann=False, imgs_dir=os.path.join(data_dir, attrs[0]), gts_dir=gts_dir, ) return dict(args=args, factory="TotaltextDataset") elif "tr400" in name: data_dir = DatasetCatalog.DATA_DIR attrs = DatasetCatalog.DATASETS[name] if len(attrs) > 1: gts_dir = os.path.join(data_dir, attrs[1]) else: gts_dir = None args = dict( use_charann=False, imgs_dir=os.path.join(data_dir, attrs[0]), gts_dir=gts_dir, ) return dict(args=args, factory="TotaltextDataset") raise RuntimeError("Dataset not available: {}".format(name)) class ModelCatalog(object): S3_C2_DETECTRON_URL = "https://dl.fbaipublicfiles.com/detectron" C2_IMAGENET_MODELS = { 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth', 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', "MSRA/R-50": "ImageNetPretrained/MSRA/R-50.pkl", "MSRA/R-50-GN": "ImageNetPretrained/47261647/R-50-GN.pkl", "MSRA/R-101": "ImageNetPretrained/MSRA/R-101.pkl", "MSRA/R-101-GN": "ImageNetPretrained/47592356/R-101-GN.pkl", "FAIR/20171220/X-101-32x8d": "ImageNetPretrained/20171220/X-101-32x8d.pkl", } C2_DETECTRON_SUFFIX = "output/train/{}coco_2014_train%3A{}coco_2014_valminusminival/generalized_rcnn/model_final.pkl" C2_DETECTRON_MODELS = { "35857197/e2e_faster_rcnn_R-50-C4_1x": "01_33_49.iAX0mXvW", "35857345/e2e_faster_rcnn_R-50-FPN_1x": "01_36_30.cUF7QR7I", "35857890/e2e_faster_rcnn_R-101-FPN_1x": "01_38_50.sNxI7sX7", "36761737/e2e_faster_rcnn_X-101-32x8d-FPN_1x": "06_31_39.5MIHi1fZ", "35858791/e2e_mask_rcnn_R-50-C4_1x": "01_45_57.ZgkA7hPB", "35858933/e2e_mask_rcnn_R-50-FPN_1x": "01_48_14.DzEQe4wC", "35861795/e2e_mask_rcnn_R-101-FPN_1x": "02_31_37.KqyEK4tT", "36761843/e2e_mask_rcnn_X-101-32x8d-FPN_1x": "06_35_59.RZotkLKI", "37129812/e2e_mask_rcnn_X-152-32x8d-FPN-IN5k_1.44x": "09_35_36.8pzTQKYK", # keypoints "37697547/e2e_keypoint_rcnn_R-50-FPN_1x": "08_42_54.kdzV35ao" } @staticmethod def get(name): if name.startswith("Caffe2Detectron/COCO"): return ModelCatalog.get_c2_detectron_12_2017_baselines(name) if name.startswith("ImageNetPretrained"): return ModelCatalog.get_c2_imagenet_pretrained(name) raise RuntimeError("model not present in the catalog {}".format(name)) @staticmethod def get_c2_imagenet_pretrained(name): prefix = ModelCatalog.S3_C2_DETECTRON_URL name = name[len("ImageNetPretrained/") :] name = ModelCatalog.C2_IMAGENET_MODELS[name] if 'resnet34' in name or 'resnet18' in name: return name url = "/".join([prefix, name]) return url @staticmethod def get_c2_detectron_12_2017_baselines(name): # Detectron C2 models are stored following the structure # prefix/<model_id>/2012_2017_baselines/<model_name>.yaml.<signature>/suffix # we use as identifiers in the catalog Caffe2Detectron/COCO/<model_id>/<model_name> prefix = ModelCatalog.S3_C2_DETECTRON_URL suffix = ModelCatalog.C2_DETECTRON_SUFFIX # remove identification prefix name = name[len("Caffe2Detectron/COCO/") :] # split in <model_id> and <model_name> model_id, model_name = name.split("/") # parsing to make it match the url address from the Caffe2 models model_name = "{}.yaml".format(model_name) signature = ModelCatalog.C2_DETECTRON_MODELS[name] unique_name = ".".join([model_name, signature]) url = "/".join([prefix, model_id, "12_2017_baselines", unique_name, suffix]) return url
11,140
45.810924
121
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/nms.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. # from ._utils import _C from maskrcnn_benchmark import _C # nms = _C.nms from apex import amp # Only valid with fp32 inputs - give AMP the hint nms = amp.float_function(_C.nms) # nms.__doc__ = """ # This function performs Non-maximum suppresion"""
323
26
71
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/batch_norm.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import torch from torch import nn class FrozenBatchNorm2d(nn.Module): """ BatchNorm2d where the batch statistics and the affine parameters are fixed """ def __init__(self, n): super(FrozenBatchNorm2d, self).__init__() self.register_buffer("weight", torch.ones(n)) self.register_buffer("bias", torch.zeros(n)) self.register_buffer("running_mean", torch.zeros(n)) self.register_buffer("running_var", torch.ones(n)) def forward(self, x): # Cast all fixed parameters to half() if necessary if x.dtype == torch.float16: self.weight = self.weight.half() self.bias = self.bias.half() self.running_mean = self.running_mean.half() self.running_var = self.running_var.half() scale = self.weight * self.running_var.rsqrt() bias = self.bias - self.running_mean * scale scale = scale.reshape(1, -1, 1, 1) bias = bias.reshape(1, -1, 1, 1) return x * scale + bias
1,093
34.290323
71
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/roi_pool.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import torch from torch import nn from torch.autograd import Function from torch.autograd.function import once_differentiable from torch.nn.modules.utils import _pair from maskrcnn_benchmark import _C from apex import amp class _ROIPool(Function): @staticmethod def forward(ctx, input, roi, output_size, spatial_scale): ctx.output_size = _pair(output_size) ctx.spatial_scale = spatial_scale ctx.input_shape = input.size() output, argmax = _C.roi_pool_forward( input, roi, spatial_scale, output_size[0], output_size[1] ) ctx.save_for_backward(input, roi, argmax) return output @staticmethod @once_differentiable def backward(ctx, grad_output): input, rois, argmax = ctx.saved_tensors output_size = ctx.output_size spatial_scale = ctx.spatial_scale bs, ch, h, w = ctx.input_shape grad_input = _C.roi_pool_backward( grad_output, input, rois, argmax, spatial_scale, output_size[0], output_size[1], bs, ch, h, w, ) return grad_input, None, None, None roi_pool = _ROIPool.apply class ROIPool(nn.Module): def __init__(self, output_size, spatial_scale): super(ROIPool, self).__init__() self.output_size = output_size self.spatial_scale = spatial_scale @amp.float_function def forward(self, input, rois): return roi_pool(input, rois, self.output_size, self.spatial_scale) def __repr__(self): tmpstr = self.__class__.__name__ + "(" tmpstr += "output_size=" + str(self.output_size) tmpstr += ", spatial_scale=" + str(self.spatial_scale) tmpstr += ")" return tmpstr
1,899
28.230769
74
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/roi_align.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import torch from torch import nn from torch.autograd import Function from torch.autograd.function import once_differentiable from torch.nn.modules.utils import _pair from maskrcnn_benchmark import _C from apex import amp class _ROIAlign(Function): @staticmethod def forward(ctx, input, roi, output_size, spatial_scale, sampling_ratio): ctx.save_for_backward(roi) ctx.output_size = _pair(output_size) ctx.spatial_scale = spatial_scale ctx.sampling_ratio = sampling_ratio ctx.input_shape = input.size() output = _C.roi_align_forward( input, roi, spatial_scale, output_size[0], output_size[1], sampling_ratio ) return output @staticmethod @once_differentiable def backward(ctx, grad_output): rois, = ctx.saved_tensors output_size = ctx.output_size spatial_scale = ctx.spatial_scale sampling_ratio = ctx.sampling_ratio bs, ch, h, w = ctx.input_shape grad_input = _C.roi_align_backward( grad_output, rois, spatial_scale, output_size[0], output_size[1], bs, ch, h, w, sampling_ratio, ) return grad_input, None, None, None, None roi_align = _ROIAlign.apply class ROIAlign(nn.Module): def __init__(self, output_size, spatial_scale, sampling_ratio): super(ROIAlign, self).__init__() self.output_size = output_size self.spatial_scale = spatial_scale self.sampling_ratio = sampling_ratio @amp.float_function def forward(self, input, rois): return roi_align( input, rois, self.output_size, self.spatial_scale, self.sampling_ratio ) def __repr__(self): tmpstr = self.__class__.__name__ + "(" tmpstr += "output_size=" + str(self.output_size) tmpstr += ", spatial_scale=" + str(self.spatial_scale) tmpstr += ", sampling_ratio=" + str(self.sampling_ratio) tmpstr += ")" return tmpstr
2,154
29.785714
85
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/smooth_l1_loss.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import torch # TODO maybe push this to nn? def smooth_l1_loss(input, target, beta=1. / 9, size_average=True): """ very similar to the smooth_l1_loss from pytorch, but with the extra beta parameter """ n = torch.abs(input - target) cond = n < beta loss = torch.where(cond, 0.5 * n ** 2 / beta, n - 0.5 * beta) if size_average: return loss.mean() return loss.sum()
481
27.352941
71
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/_utils.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import glob import os.path import torch try: from torch.utils.cpp_extension import load as load_ext from torch.utils.cpp_extension import CUDA_HOME except ImportError: raise ImportError("The cpp layer extensions requires PyTorch 0.4 or higher") def _load_C_extensions(): this_dir = os.path.dirname(os.path.abspath(__file__)) this_dir = os.path.dirname(this_dir) this_dir = os.path.join(this_dir, "csrc") main_file = glob.glob(os.path.join(this_dir, "*.cpp")) source_cpu = glob.glob(os.path.join(this_dir, "cpu", "*.cpp")) source_cuda = glob.glob(os.path.join(this_dir, "cuda", "*.cu")) source = main_file + source_cpu extra_cflags = [] if torch.cuda.is_available() and CUDA_HOME is not None: source.extend(source_cuda) extra_cflags = ["-DWITH_CUDA"] source = [os.path.join(this_dir, s) for s in source] extra_include_paths = [this_dir] return load_ext( "torchvision", source, extra_cflags=extra_cflags, extra_include_paths=extra_include_paths, ) _C = _load_C_extensions()
1,165
28.15
80
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/misc.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. """ helper class that supports empty tensors on some nn functions. Ideally, add support directly in PyTorch to empty tensors in those functions. This can be removed once https://github.com/pytorch/pytorch/issues/12013 is implemented """ import math import torch from torch import nn from torch.nn.modules.utils import _ntuple class _NewEmptyTensorOp(torch.autograd.Function): @staticmethod def forward(ctx, x, new_shape): ctx.shape = x.shape return x.new_empty(new_shape) @staticmethod def backward(ctx, grad): shape = ctx.shape return _NewEmptyTensorOp.apply(grad, shape), None class Conv2d(torch.nn.Conv2d): def forward(self, x): if x.numel() > 0: return super(Conv2d, self).forward(x) # get output shape output_shape = [ (i + 2 * p - (di * (k - 1) + 1)) // d + 1 for i, p, di, k, d in zip( x.shape[-2:], self.padding, self.dilation, self.kernel_size, self.stride ) ] output_shape = [x.shape[0], self.weight.shape[0]] + output_shape return _NewEmptyTensorOp.apply(x, output_shape) class ConvTranspose2d(torch.nn.ConvTranspose2d): def forward(self, x): if x.numel() > 0: return super(ConvTranspose2d, self).forward(x) # get output shape output_shape = [ (i - 1) * d - 2 * p + (di * (k - 1) + 1) + op for i, p, di, k, d, op in zip( x.shape[-2:], self.padding, self.dilation, self.kernel_size, self.stride, self.output_padding, ) ] output_shape = [x.shape[0], self.bias.shape[0]] + output_shape return _NewEmptyTensorOp.apply(x, output_shape) def interpolate( input, size=None, scale_factor=None, mode="nearest", align_corners=None ): if input.numel() > 0: return torch.nn.functional.interpolate( input, size, scale_factor, mode, align_corners ) def _check_size_scale_factor(dim): if size is None and scale_factor is None: raise ValueError("either size or scale_factor should be defined") if size is not None and scale_factor is not None: raise ValueError("only one of size or scale_factor should be defined") if ( scale_factor is not None and isinstance(scale_factor, tuple) and len(scale_factor) != dim ): raise ValueError( "scale_factor shape must match input shape. " "Input is {}D, scale_factor size is {}".format(dim, len(scale_factor)) ) def _output_size(dim): _check_size_scale_factor(dim) if size is not None: return size scale_factors = _ntuple(dim)(scale_factor) # math.floor might return float in py2.7 return [ int(math.floor(input.size(i + 2) * scale_factors[i])) for i in range(dim) ] output_shape = tuple(_output_size(2)) output_shape = input.shape[:-2] + output_shape return _NewEmptyTensorOp.apply(input, output_shape) class DFConv2d(nn.Module): """Deformable convolutional layer""" def __init__( self, in_channels, out_channels, with_modulated_dcn=True, kernel_size=3, stride=1, groups=1, dilation=1, deformable_groups=1, bias=False ): super(DFConv2d, self).__init__() if isinstance(kernel_size, (list, tuple)): assert len(kernel_size) == 2 offset_base_channels = kernel_size[0] * kernel_size[1] else: offset_base_channels = kernel_size * kernel_size if with_modulated_dcn: from maskrcnn_benchmark.layers import ModulatedDeformConv offset_channels = offset_base_channels * 3 #default: 27 conv_block = ModulatedDeformConv else: from maskrcnn_benchmark.layers import DeformConv offset_channels = offset_base_channels * 2 #default: 18 conv_block = DeformConv self.offset = Conv2d( in_channels, deformable_groups * offset_channels, kernel_size=kernel_size, stride= stride, padding= dilation, groups=1, dilation=dilation ) for l in [self.offset,]: nn.init.kaiming_uniform_(l.weight, a=1) torch.nn.init.constant_(l.bias, 0.) self.conv = conv_block( in_channels, out_channels, kernel_size=kernel_size, stride= stride, padding=dilation, dilation=dilation, groups=groups, deformable_groups=deformable_groups, bias=bias ) self.with_modulated_dcn = with_modulated_dcn self.kernel_size = kernel_size self.stride = stride self.padding = dilation self.dilation = dilation def forward(self, x): if x.numel() > 0: if not self.with_modulated_dcn: offset = self.offset(x) x = self.conv(x, offset) else: offset_mask = self.offset(x) offset = offset_mask[:, :18, :, :] mask = offset_mask[:, -9:, :, :].sigmoid() x = self.conv(x, offset, mask) return x # get output shape output_shape = [ (i + 2 * p - (di * (k - 1) + 1)) // d + 1 for i, p, di, k, d in zip( x.shape[-2:], self.padding, self.dilation, self.kernel_size, self.stride ) ] output_shape = [x.shape[0], self.conv.weight.shape[0]] + output_shape return _NewEmptyTensorOp.apply(x, output_shape)
6,035
31.451613
88
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/__init__.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import torch from .batch_norm import FrozenBatchNorm2d from .misc import Conv2d from .misc import DFConv2d from .misc import ConvTranspose2d from .misc import interpolate from .nms import nms from .roi_align import ROIAlign from .roi_align import roi_align from .roi_pool import ROIPool from .roi_pool import roi_pool from .smooth_l1_loss import smooth_l1_loss from .dcn.deform_conv_func import deform_conv, modulated_deform_conv from .dcn.deform_conv_module import DeformConv, ModulatedDeformConv, ModulatedDeformConvPack from .dcn.deform_pool_func import deform_roi_pooling from .dcn.deform_pool_module import DeformRoIPooling, DeformRoIPoolingPack, ModulatedDeformRoIPoolingPack __all__ = [ "nms", "roi_align", "ROIAlign", "roi_pool", "ROIPool", "smooth_l1_loss", "Conv2d", "DFConv2d", "ConvTranspose2d", "interpolate", "BatchNorm2d", "FrozenBatchNorm2d", 'deform_conv', 'modulated_deform_conv', 'DeformConv', 'ModulatedDeformConv', 'ModulatedDeformConvPack', 'deform_roi_pooling', 'DeformRoIPooling', 'DeformRoIPoolingPack', 'ModulatedDeformRoIPoolingPack', ] # __all__ = ["nms", "roi_align", "ROIAlign", "roi_pool", "ROIPool", "smooth_l1_loss", "Conv2d", "ConvTranspose2d", "interpolate", "FrozenBatchNorm2d"]
1,372
30.930233
150
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/dcn/deform_conv_func.py
import torch from torch.autograd import Function from torch.autograd.function import once_differentiable from torch.nn.modules.utils import _pair from maskrcnn_benchmark import _C class DeformConvFunction(Function): @staticmethod def forward( ctx, input, offset, weight, stride=1, padding=0, dilation=1, groups=1, deformable_groups=1, im2col_step=64 ): if input is not None and input.dim() != 4: raise ValueError( "Expected 4D tensor as input, got {}D tensor instead.".format( input.dim())) ctx.stride = _pair(stride) ctx.padding = _pair(padding) ctx.dilation = _pair(dilation) ctx.groups = groups ctx.deformable_groups = deformable_groups ctx.im2col_step = im2col_step ctx.save_for_backward(input, offset, weight) output = input.new_empty( DeformConvFunction._output_size(input, weight, ctx.padding, ctx.dilation, ctx.stride)) ctx.bufs_ = [input.new_empty(0), input.new_empty(0)] # columns, ones if not input.is_cuda: raise NotImplementedError else: cur_im2col_step = min(ctx.im2col_step, input.shape[0]) assert (input.shape[0] % cur_im2col_step) == 0, 'im2col step must divide batchsize' _C.deform_conv_forward( input, weight, offset, output, ctx.bufs_[0], ctx.bufs_[1], weight.size(3), weight.size(2), ctx.stride[1], ctx.stride[0], ctx.padding[1], ctx.padding[0], ctx.dilation[1], ctx.dilation[0], ctx.groups, ctx.deformable_groups, cur_im2col_step ) return output @staticmethod @once_differentiable def backward(ctx, grad_output): input, offset, weight = ctx.saved_tensors grad_input = grad_offset = grad_weight = None if not grad_output.is_cuda: raise NotImplementedError else: cur_im2col_step = min(ctx.im2col_step, input.shape[0]) assert (input.shape[0] % cur_im2col_step) == 0, 'im2col step must divide batchsize' if ctx.needs_input_grad[0] or ctx.needs_input_grad[1]: grad_input = torch.zeros_like(input) grad_offset = torch.zeros_like(offset) _C.deform_conv_backward_input( input, offset, grad_output, grad_input, grad_offset, weight, ctx.bufs_[0], weight.size(3), weight.size(2), ctx.stride[1], ctx.stride[0], ctx.padding[1], ctx.padding[0], ctx.dilation[1], ctx.dilation[0], ctx.groups, ctx.deformable_groups, cur_im2col_step ) if ctx.needs_input_grad[2]: grad_weight = torch.zeros_like(weight) _C.deform_conv_backward_parameters( input, offset, grad_output, grad_weight, ctx.bufs_[0], ctx.bufs_[1], weight.size(3), weight.size(2), ctx.stride[1], ctx.stride[0], ctx.padding[1], ctx.padding[0], ctx.dilation[1], ctx.dilation[0], ctx.groups, ctx.deformable_groups, 1, cur_im2col_step ) return (grad_input, grad_offset, grad_weight, None, None, None, None, None) @staticmethod def _output_size(input, weight, padding, dilation, stride): channels = weight.size(0) output_size = (input.size(0), channels) for d in range(input.dim() - 2): in_size = input.size(d + 2) pad = padding[d] kernel = dilation[d] * (weight.size(d + 2) - 1) + 1 stride_ = stride[d] output_size += ((in_size + (2 * pad) - kernel) // stride_ + 1, ) if not all(map(lambda s: s > 0, output_size)): raise ValueError( "convolution input is too small (output would be {})".format( 'x'.join(map(str, output_size)))) return output_size class ModulatedDeformConvFunction(Function): @staticmethod def forward( ctx, input, offset, mask, weight, bias=None, stride=1, padding=0, dilation=1, groups=1, deformable_groups=1 ): ctx.stride = stride ctx.padding = padding ctx.dilation = dilation ctx.groups = groups ctx.deformable_groups = deformable_groups ctx.with_bias = bias is not None if not ctx.with_bias: bias = input.new_empty(1) # fake tensor if not input.is_cuda: raise NotImplementedError if weight.requires_grad or mask.requires_grad or offset.requires_grad \ or input.requires_grad: ctx.save_for_backward(input, offset, mask, weight, bias) output = input.new_empty( ModulatedDeformConvFunction._infer_shape(ctx, input, weight)) ctx._bufs = [input.new_empty(0), input.new_empty(0)] _C.modulated_deform_conv_forward( input, weight, bias, ctx._bufs[0], offset, mask, output, ctx._bufs[1], weight.shape[2], weight.shape[3], ctx.stride, ctx.stride, ctx.padding, ctx.padding, ctx.dilation, ctx.dilation, ctx.groups, ctx.deformable_groups, ctx.with_bias ) return output @staticmethod @once_differentiable def backward(ctx, grad_output): if not grad_output.is_cuda: raise NotImplementedError input, offset, mask, weight, bias = ctx.saved_tensors grad_input = torch.zeros_like(input) grad_offset = torch.zeros_like(offset) grad_mask = torch.zeros_like(mask) grad_weight = torch.zeros_like(weight) grad_bias = torch.zeros_like(bias) _C.modulated_deform_conv_backward( input, weight, bias, ctx._bufs[0], offset, mask, ctx._bufs[1], grad_input, grad_weight, grad_bias, grad_offset, grad_mask, grad_output, weight.shape[2], weight.shape[3], ctx.stride, ctx.stride, ctx.padding, ctx.padding, ctx.dilation, ctx.dilation, ctx.groups, ctx.deformable_groups, ctx.with_bias ) if not ctx.with_bias: grad_bias = None return (grad_input, grad_offset, grad_mask, grad_weight, grad_bias, None, None, None, None, None) @staticmethod def _infer_shape(ctx, input, weight): n = input.size(0) channels_out = weight.size(0) height, width = input.shape[2:4] kernel_h, kernel_w = weight.shape[2:4] height_out = (height + 2 * ctx.padding - (ctx.dilation * (kernel_h - 1) + 1)) // ctx.stride + 1 width_out = (width + 2 * ctx.padding - (ctx.dilation * (kernel_w - 1) + 1)) // ctx.stride + 1 return n, channels_out, height_out, width_out deform_conv = DeformConvFunction.apply modulated_deform_conv = ModulatedDeformConvFunction.apply
8,309
30.596958
83
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/dcn/deform_pool_func.py
import torch from torch.autograd import Function from torch.autograd.function import once_differentiable from maskrcnn_benchmark import _C class DeformRoIPoolingFunction(Function): @staticmethod def forward( ctx, data, rois, offset, spatial_scale, out_size, out_channels, no_trans, group_size=1, part_size=None, sample_per_part=4, trans_std=.0 ): ctx.spatial_scale = spatial_scale ctx.out_size = out_size ctx.out_channels = out_channels ctx.no_trans = no_trans ctx.group_size = group_size ctx.part_size = out_size if part_size is None else part_size ctx.sample_per_part = sample_per_part ctx.trans_std = trans_std assert 0.0 <= ctx.trans_std <= 1.0 if not data.is_cuda: raise NotImplementedError n = rois.shape[0] output = data.new_empty(n, out_channels, out_size, out_size) output_count = data.new_empty(n, out_channels, out_size, out_size) _C.deform_psroi_pooling_forward( data, rois, offset, output, output_count, ctx.no_trans, ctx.spatial_scale, ctx.out_channels, ctx.group_size, ctx.out_size, ctx.part_size, ctx.sample_per_part, ctx.trans_std ) if data.requires_grad or rois.requires_grad or offset.requires_grad: ctx.save_for_backward(data, rois, offset) ctx.output_count = output_count return output @staticmethod @once_differentiable def backward(ctx, grad_output): if not grad_output.is_cuda: raise NotImplementedError data, rois, offset = ctx.saved_tensors output_count = ctx.output_count grad_input = torch.zeros_like(data) grad_rois = None grad_offset = torch.zeros_like(offset) _C.deform_psroi_pooling_backward( grad_output, data, rois, offset, output_count, grad_input, grad_offset, ctx.no_trans, ctx.spatial_scale, ctx.out_channels, ctx.group_size, ctx.out_size, ctx.part_size, ctx.sample_per_part, ctx.trans_std ) return (grad_input, grad_rois, grad_offset, None, None, None, None, None, None, None, None) deform_roi_pooling = DeformRoIPoolingFunction.apply
2,595
26.041667
99
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/dcn/deform_pool_module.py
from torch import nn from .deform_pool_func import deform_roi_pooling class DeformRoIPooling(nn.Module): def __init__(self, spatial_scale, out_size, out_channels, no_trans, group_size=1, part_size=None, sample_per_part=4, trans_std=.0): super(DeformRoIPooling, self).__init__() self.spatial_scale = spatial_scale self.out_size = out_size self.out_channels = out_channels self.no_trans = no_trans self.group_size = group_size self.part_size = out_size if part_size is None else part_size self.sample_per_part = sample_per_part self.trans_std = trans_std def forward(self, data, rois, offset): if self.no_trans: offset = data.new_empty(0) return deform_roi_pooling( data, rois, offset, self.spatial_scale, self.out_size, self.out_channels, self.no_trans, self.group_size, self.part_size, self.sample_per_part, self.trans_std) class DeformRoIPoolingPack(DeformRoIPooling): def __init__(self, spatial_scale, out_size, out_channels, no_trans, group_size=1, part_size=None, sample_per_part=4, trans_std=.0, deform_fc_channels=1024): super(DeformRoIPoolingPack, self).__init__(spatial_scale, out_size, out_channels, no_trans, group_size, part_size, sample_per_part, trans_std) self.deform_fc_channels = deform_fc_channels if not no_trans: self.offset_fc = nn.Sequential( nn.Linear(self.out_size * self.out_size * self.out_channels, self.deform_fc_channels), nn.ReLU(inplace=True), nn.Linear(self.deform_fc_channels, self.deform_fc_channels), nn.ReLU(inplace=True), nn.Linear(self.deform_fc_channels, self.out_size * self.out_size * 2)) self.offset_fc[-1].weight.data.zero_() self.offset_fc[-1].bias.data.zero_() def forward(self, data, rois): assert data.size(1) == self.out_channels if self.no_trans: offset = data.new_empty(0) return deform_roi_pooling( data, rois, offset, self.spatial_scale, self.out_size, self.out_channels, self.no_trans, self.group_size, self.part_size, self.sample_per_part, self.trans_std) else: n = rois.shape[0] offset = data.new_empty(0) x = deform_roi_pooling(data, rois, offset, self.spatial_scale, self.out_size, self.out_channels, True, self.group_size, self.part_size, self.sample_per_part, self.trans_std) offset = self.offset_fc(x.view(n, -1)) offset = offset.view(n, 2, self.out_size, self.out_size) return deform_roi_pooling( data, rois, offset, self.spatial_scale, self.out_size, self.out_channels, self.no_trans, self.group_size, self.part_size, self.sample_per_part, self.trans_std) class ModulatedDeformRoIPoolingPack(DeformRoIPooling): def __init__(self, spatial_scale, out_size, out_channels, no_trans, group_size=1, part_size=None, sample_per_part=4, trans_std=.0, deform_fc_channels=1024): super(ModulatedDeformRoIPoolingPack, self).__init__( spatial_scale, out_size, out_channels, no_trans, group_size, part_size, sample_per_part, trans_std) self.deform_fc_channels = deform_fc_channels if not no_trans: self.offset_fc = nn.Sequential( nn.Linear(self.out_size * self.out_size * self.out_channels, self.deform_fc_channels), nn.ReLU(inplace=True), nn.Linear(self.deform_fc_channels, self.deform_fc_channels), nn.ReLU(inplace=True), nn.Linear(self.deform_fc_channels, self.out_size * self.out_size * 2)) self.offset_fc[-1].weight.data.zero_() self.offset_fc[-1].bias.data.zero_() self.mask_fc = nn.Sequential( nn.Linear(self.out_size * self.out_size * self.out_channels, self.deform_fc_channels), nn.ReLU(inplace=True), nn.Linear(self.deform_fc_channels, self.out_size * self.out_size * 1), nn.Sigmoid()) self.mask_fc[2].weight.data.zero_() self.mask_fc[2].bias.data.zero_() def forward(self, data, rois): assert data.size(1) == self.out_channels if self.no_trans: offset = data.new_empty(0) return deform_roi_pooling( data, rois, offset, self.spatial_scale, self.out_size, self.out_channels, self.no_trans, self.group_size, self.part_size, self.sample_per_part, self.trans_std) else: n = rois.shape[0] offset = data.new_empty(0) x = deform_roi_pooling(data, rois, offset, self.spatial_scale, self.out_size, self.out_channels, True, self.group_size, self.part_size, self.sample_per_part, self.trans_std) offset = self.offset_fc(x.view(n, -1)) offset = offset.view(n, 2, self.out_size, self.out_size) mask = self.mask_fc(x.view(n, -1)) mask = mask.view(n, 1, self.out_size, self.out_size) return deform_roi_pooling( data, rois, offset, self.spatial_scale, self.out_size, self.out_channels, self.no_trans, self.group_size, self.part_size, self.sample_per_part, self.trans_std) * mask
6,307
40.774834
79
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/dcn/__init__.py
# # Copied From [mmdetection](https://github.com/open-mmlab/mmdetection/tree/master/mmdet/ops/dcn) #
101
24.5
96
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/layers/dcn/deform_conv_module.py
import math import torch import torch.nn as nn from torch.nn.modules.utils import _pair from .deform_conv_func import deform_conv, modulated_deform_conv class DeformConv(nn.Module): def __init__( self, in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, deformable_groups=1, bias=False ): assert not bias super(DeformConv, self).__init__() self.with_bias = bias assert in_channels % groups == 0, \ 'in_channels {} cannot be divisible by groups {}'.format( in_channels, groups) assert out_channels % groups == 0, \ 'out_channels {} cannot be divisible by groups {}'.format( out_channels, groups) self.in_channels = in_channels self.out_channels = out_channels self.kernel_size = _pair(kernel_size) self.stride = _pair(stride) self.padding = _pair(padding) self.dilation = _pair(dilation) self.groups = groups self.deformable_groups = deformable_groups self.weight = nn.Parameter( torch.Tensor(out_channels, in_channels // self.groups, *self.kernel_size)) self.reset_parameters() def reset_parameters(self): n = self.in_channels for k in self.kernel_size: n *= k stdv = 1. / math.sqrt(n) self.weight.data.uniform_(-stdv, stdv) def forward(self, input, offset): return deform_conv(input, offset, self.weight, self.stride, self.padding, self.dilation, self.groups, self.deformable_groups) def __repr__(self): return "".join([ "{}(".format(self.__class__.__name__), "in_channels={}, ".format(self.in_channels), "out_channels={}, ".format(self.out_channels), "kernel_size={}, ".format(self.kernel_size), "stride={}, ".format(self.stride), "dilation={}, ".format(self.dilation), "padding={}, ".format(self.padding), "groups={}, ".format(self.groups), "deformable_groups={}, ".format(self.deformable_groups), "bias={})".format(self.with_bias), ]) class ModulatedDeformConv(nn.Module): def __init__( self, in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, deformable_groups=1, bias=True ): super(ModulatedDeformConv, self).__init__() self.in_channels = in_channels self.out_channels = out_channels self.kernel_size = _pair(kernel_size) self.stride = stride self.padding = padding self.dilation = dilation self.groups = groups self.deformable_groups = deformable_groups self.with_bias = bias self.weight = nn.Parameter(torch.Tensor( out_channels, in_channels // groups, *self.kernel_size )) if bias: self.bias = nn.Parameter(torch.Tensor(out_channels)) else: self.register_parameter('bias', None) self.reset_parameters() def reset_parameters(self): n = self.in_channels for k in self.kernel_size: n *= k stdv = 1. / math.sqrt(n) self.weight.data.uniform_(-stdv, stdv) if self.bias is not None: self.bias.data.zero_() def forward(self, input, offset, mask): return modulated_deform_conv( input, offset, mask, self.weight, self.bias, self.stride, self.padding, self.dilation, self.groups, self.deformable_groups) def __repr__(self): return "".join([ "{}(".format(self.__class__.__name__), "in_channels={}, ".format(self.in_channels), "out_channels={}, ".format(self.out_channels), "kernel_size={}, ".format(self.kernel_size), "stride={}, ".format(self.stride), "dilation={}, ".format(self.dilation), "padding={}, ".format(self.padding), "groups={}, ".format(self.groups), "deformable_groups={}, ".format(self.deformable_groups), "bias={})".format(self.with_bias), ]) class ModulatedDeformConvPack(ModulatedDeformConv): def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, deformable_groups=1, bias=True): super(ModulatedDeformConvPack, self).__init__( in_channels, out_channels, kernel_size, stride, padding, dilation, groups, deformable_groups, bias) self.conv_offset_mask = nn.Conv2d( self.in_channels // self.groups, self.deformable_groups * 3 * self.kernel_size[0] * self.kernel_size[1], kernel_size=self.kernel_size, stride=_pair(self.stride), padding=_pair(self.padding), bias=True) self.init_offset() def init_offset(self): self.conv_offset_mask.weight.data.zero_() self.conv_offset_mask.bias.data.zero_() def forward(self, input): out = self.conv_offset_mask(input) o1, o2, mask = torch.chunk(out, 3, dim=1) offset = torch.cat((o1, o2), dim=1) mask = torch.sigmoid(mask) return modulated_deform_conv( input, offset, mask, self.weight, self.bias, self.stride, self.padding, self.dilation, self.groups, self.deformable_groups)
5,802
31.601124
78
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/engine/text_inference.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import datetime import logging import os import pickle import subprocess import time import cv2 import numpy as np import torch from maskrcnn_benchmark.utils.chars import char2num, get_tight_rect, getstr_grid from PIL import Image, ImageDraw from tqdm import tqdm from ..utils.comm import is_main_process, scatter_gather, synchronize import pdb # TO DO: format output with dictionnary def compute_on_dataset(model, data_loader, device, cfg): model.eval() results_dict = {} seg_results = [] cpu_device = torch.device("cpu") total_time = 0 for _, batch in tqdm(enumerate(data_loader)): images, targets, image_paths = batch images = images.to(device) with torch.no_grad(): if cfg.MODEL.SEG_ON: predictions, proposals, seg_results_dict = model( images ) seg_results.append( [image_paths, proposals, seg_results_dict['rotated_boxes'], seg_results_dict['polygons'], seg_results_dict['preds'], seg_results_dict['scores']] ) # if cfg.MODEL.MASK_ON and predictions is not None: if predictions is not None: if cfg.MODEL.CHAR_MASK_ON or cfg.SEQUENCE.SEQ_ON: global_predictions = predictions[0] char_predictions = predictions[1] char_mask = char_predictions["char_mask"] boxes = char_predictions["boxes"] seq_words = char_predictions["seq_outputs"] seq_scores = char_predictions["seq_scores"] detailed_seq_scores = char_predictions["detailed_seq_scores"] global_predictions = [o.to(cpu_device) for o in global_predictions] results_dict.update( { image_paths[0]: [ global_predictions[0], char_mask, boxes, seq_words, seq_scores, detailed_seq_scores, ] } ) else: global_predictions = [o.to(cpu_device) for o in predictions] results_dict.update( { image_paths[0]: [ global_predictions[0], ] } ) else: predictions = model(images) if predictions is not None: if not (cfg.MODEL.CHAR_MASK_ON and cfg.SEQUENCE.SEQ_ON): global_predictions = predictions global_predictions = [o.to(cpu_device) for o in global_predictions] results_dict.update( { image_paths[0]: [ global_predictions[0], ] } ) else: global_predictions = predictions[0] char_predictions = predictions[1] if cfg.MODEL.CHAR_MASK_ON: char_mask = char_predictions["char_mask"] else: char_mask = None boxes = char_predictions["boxes"] seq_words = char_predictions["seq_outputs"] seq_scores = char_predictions["seq_scores"] detailed_seq_scores = char_predictions["detailed_seq_scores"] global_predictions = [o.to(cpu_device) for o in global_predictions] results_dict.update( { image_paths[0]: [ global_predictions[0], char_mask, boxes, seq_words, seq_scores, detailed_seq_scores, ] } ) return results_dict, seg_results def polygon2rbox(polygon, image_height, image_width): poly = np.array(polygon).reshape((-1, 2)) rect = cv2.minAreaRect(poly) corners = cv2.boxPoints(rect) corners = np.array(corners, dtype="int") pts = get_tight_rect(corners, 0, 0, image_height, image_width, 1) pts = list(map(int, pts)) return pts def mask2polygon(mask, box, im_size, threshold=0.5, output_folder=None): # mask 32*128 image_width, image_height = im_size[0], im_size[1] box_h = box[3] - box[1] box_w = box[2] - box[0] cls_polys = (mask * 255).astype(np.uint8) poly_map = np.array(Image.fromarray(cls_polys).resize((box_w, box_h))) poly_map = poly_map.astype(np.float32) / 255 poly_map = cv2.GaussianBlur(poly_map, (3, 3), sigmaX=3) ret, poly_map = cv2.threshold(poly_map, threshold, 1, cv2.THRESH_BINARY) if "total_text" in output_folder or "cute80" in output_folder: SE1 = cv2.getStructuringElement(cv2.MORPH_RECT, (3, 3)) poly_map = cv2.erode(poly_map, SE1) poly_map = cv2.dilate(poly_map, SE1) poly_map = cv2.morphologyEx(poly_map, cv2.MORPH_CLOSE, SE1) try: _, contours, _ = cv2.findContours( (poly_map * 255).astype(np.uint8), cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE ) except: contours, _ = cv2.findContours( (poly_map * 255).astype(np.uint8), cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE ) if len(contours) == 0: # print(contours) # print(len(contours)) return None max_area = 0 max_cnt = contours[0] for cnt in contours: area = cv2.contourArea(cnt) if area > max_area: max_area = area max_cnt = cnt # perimeter = cv2.arcLength(max_cnt, True) epsilon = 0.01 * cv2.arcLength(max_cnt, True) approx = cv2.approxPolyDP(max_cnt, epsilon, True) pts = approx.reshape((-1, 2)) pts[:, 0] = pts[:, 0] + box[0] pts[:, 1] = pts[:, 1] + box[1] polygon = list(pts.reshape((-1,))) polygon = list(map(int, polygon)) if len(polygon) < 6: return None else: SE1 = cv2.getStructuringElement(cv2.MORPH_RECT, (3, 3)) poly_map = cv2.erode(poly_map, SE1) poly_map = cv2.dilate(poly_map, SE1) poly_map = cv2.morphologyEx(poly_map, cv2.MORPH_CLOSE, SE1) idy, idx = np.where(poly_map == 1) xy = np.vstack((idx, idy)) xy = np.transpose(xy) hull = cv2.convexHull(xy, clockwise=True) # reverse order of points. if hull is None: return None hull = hull[::-1] # find minimum area bounding box. rect = cv2.minAreaRect(hull) corners = cv2.boxPoints(rect) corners = np.array(corners, dtype="int") pts = get_tight_rect(corners, box[0], box[1], image_height, image_width, 1) polygon = [x * 1.0 for x in pts] polygon = list(map(int, polygon)) return polygon def _accumulate_predictions_from_multiple_gpus(predictions_per_gpu): all_predictions = scatter_gather(predictions_per_gpu) if not is_main_process(): return # merge the list of dicts predictions = {} for p in all_predictions: predictions.update(p) return predictions def format_output(out_dir, boxes, img_name): with open( os.path.join(out_dir, "res_" + img_name.split(".")[0] + ".txt"), "wt" ) as res: ## char score save dir ssur_name = os.path.join(out_dir, "res_" + img_name.split(".")[0]) for i, box in enumerate(boxes): save_name = ssur_name + "_" + str(i) + ".pkl" save_dict = {} if "total_text" in out_dir or "cute80" in out_dir: # np.save(save_name, box[-2]) save_dict["seg_char_scores"] = box[-3] save_dict["seq_char_scores"] = box[-2] box = ( ",".join([str(x) for x in box[:4]]) + ";" + ",".join([str(x) for x in box[4 : 4 + int(box[-1])]]) + ";" + ",".join([str(x) for x in box[4 + int(box[-1]) : -3]]) + "," + save_name ) else: save_dict["seg_char_scores"] = box[-2] save_dict["seq_char_scores"] = box[-1] np.save(save_name, box[-1]) box = ",".join([str(x) for x in box[:-2]]) + "," + save_name with open(save_name, "wb") as f: pickle.dump(save_dict, f, protocol=2) res.write(box + "\n") def format_seg_output(results_dir, rotated_boxes_this_image, polygons_this_image, scores, img_name, ratio): height_ratio, width_ratio = ratio with open( os.path.join(results_dir, "res_" + img_name.split(".")[0] + ".txt"), "wt" ) as res: if "total_text" in results_dir or "cute80" in results_dir: for i, box in enumerate(polygons_this_image): box = box[0] box[0::2] = box[0::2] * width_ratio box[1::2] = box[1::2] * height_ratio save_dict = {} # result = ",".join([str(int(x[0])) + ',' +str(int(x[1])) for x in box]) result = ",".join([str(int(x)) for x in box]) score = scores[i].item() res.write(result + ',' + str(score) + "\n") else: for i, box in enumerate(rotated_boxes_this_image): box[0::2] = box[0::2] * width_ratio box[1::2] = box[1::2] * height_ratio save_dict = {} result = ",".join([str(int(x[0])) + ',' +str(int(x[1])) for x in box]) score = scores[i].item() res.write(result + ',' + str(score) + "\n") def process_char_mask(char_masks, boxes, threshold=192): texts, rec_scores, rec_char_scores, char_polygons = [], [], [], [] for index in range(char_masks.shape[0]): box = list(boxes[index]) box = list(map(int, box)) text, rec_score, rec_char_score, char_polygon = getstr_grid( char_masks[index, :, :, :].copy(), box, threshold=threshold ) texts.append(text) rec_scores.append(rec_score) rec_char_scores.append(rec_char_score) char_polygons.append(char_polygon) # segmss.append(segms) return texts, rec_scores, rec_char_scores, char_polygons def creat_color_map(n_class, width): splits = int(np.ceil(np.power((n_class * 1.0), 1.0 / 3))) maps = [] for i in range(splits): r = int(i * width * 1.0 / (splits - 1)) for j in range(splits): g = int(j * width * 1.0 / (splits - 1)) for k in range(splits - 1): b = int(k * width * 1.0 / (splits - 1)) maps.append((r, g, b, 200)) return maps def visualization(image, polygons, resize_ratio, colors, char_polygons=None, words=None): draw = ImageDraw.Draw(image, "RGBA") for polygon in polygons: # draw.polygon(polygon, fill=None, outline=(0, 255, 0, 255)) # print(polygon) polygon.append(polygon[0]) polygon.append(polygon[1]) # print(polygon) color = '#33FF33' draw.line(polygon, fill=color, width=5) # if char_polygons is not None: # for i, char_polygon in enumerate(char_polygons): # for j, polygon in enumerate(char_polygon): # polygon = [int(x * resize_ratio) for x in polygon] # char = words[i][j] # color = colors[char2num(char)] # draw.polygon(polygon, fill=color, outline=color) def vis_seg_map(image_path, seg_map, rotated_boxes, polygons_this_image, proposals, vis_dir): img_name = image_path.split("/")[-1] image = cv2.imread(image_path) height, width, _ = image.shape seg_map = seg_map.data.cpu().numpy() img = Image.fromarray(image).convert("RGB") # height_ratio = height / seg_map.shape[1] # width_ratio = width / seg_map.shape[2] # print('seg_map.shape:', seg_map.shape) # print('image.shape:', image.shape) seg_image = ( Image.fromarray((seg_map[0, :proposals.size[1], :proposals.size[0]] * 255).astype(np.uint8)) .convert("RGB") .resize((width, height)) ) visu_image = Image.blend(seg_image, img, 0.5) img_draw = ImageDraw.Draw(visu_image) if "total_text" in vis_dir or "cute80" in vis_dir: for box in polygons_this_image: # box[:, 0] = box[:, 0] # box[:, 1] = box[:, 1] tuple_box = [tuple(x) for x in box[0].reshape(-1, 2).tolist()] tuple_box.append(tuple_box[0]) img_draw.line(tuple_box, fill=(0, 255, 0), width=5) else: for box in rotated_boxes: # box[:, 0] = box[:, 0] # box[:, 1] = box[:, 1] tuple_box = [tuple(x) for x in box.tolist()] tuple_box.append(tuple_box[0]) img_draw.line(tuple_box, fill=(0, 255, 0), width=5) visu_image.save(vis_dir + "/seg_" + img_name) def prepare_results_for_evaluation( predictions, output_folder, model_name, seg_predictions=None, vis=False, cfg=None ): results_dir = os.path.join(output_folder, model_name + "_results") if not os.path.isdir(results_dir): os.mkdir(results_dir) seg_results_dir = os.path.join(output_folder, model_name + "_seg_results") if not os.path.isdir(seg_results_dir): os.mkdir(seg_results_dir) if vis: visu_dir = os.path.join(output_folder, model_name + "_visu") if not os.path.isdir(visu_dir): os.mkdir(visu_dir) seg_visu_dir = os.path.join(output_folder, model_name + "_seg_visu") if not os.path.isdir(seg_visu_dir): os.mkdir(seg_visu_dir) if len(seg_predictions) > 0: for seg_prediction in seg_predictions: image_paths, proposals, rotated_boxes, polygons, seg_maps, seg_scores = ( seg_prediction[0], seg_prediction[1], seg_prediction[2], seg_prediction[3], seg_prediction[4], seg_prediction[5], ) for batch_id in range(len(image_paths)): image_path = image_paths[batch_id] im_name = image_path.split("/")[-1] image = cv2.imread(image_path) height, width, _ = image.shape rotated_boxes_this_image = rotated_boxes[batch_id] polygons_this_image = polygons[batch_id] proposals_this_image = proposals[batch_id] seg_map = seg_maps[batch_id] seg_score = seg_scores[batch_id] height, width, _ = image.shape height_ratio = height / proposals_this_image.size[1] width_ratio = width / proposals_this_image.size[0] format_seg_output(seg_results_dir, rotated_boxes_this_image, polygons_this_image, seg_score, im_name, (height_ratio, width_ratio)) if vis: vis_seg_map(image_path, seg_map, rotated_boxes_this_image, polygons_this_image, proposals_this_image, seg_visu_dir) if (not cfg.MODEL.TRAIN_DETECTION_ONLY): for image_path, prediction in predictions.items(): im_name = image_path.split("/")[-1] if cfg.MODEL.CHAR_MASK_ON or cfg.SEQUENCE.SEQ_ON: global_prediction, char_mask, boxes_char, seq_words, seq_scores, detailed_seq_scores = ( prediction[0], prediction[1], prediction[2], prediction[3], prediction[4], prediction[5], ) if char_mask is not None: words, rec_scores, rec_char_scoress, char_polygons = process_char_mask( char_mask, boxes_char ) else: global_prediction = prediction[0] test_image_width, test_image_height = global_prediction.size img = Image.open(image_path) width, height = img.size resize_ratio = float(height) / test_image_height global_prediction = global_prediction.resize((width, height)) boxes = global_prediction.bbox.tolist() if cfg.MODEL.ROI_BOX_HEAD.INFERENCE_USE_BOX: scores = global_prediction.get_field("scores").tolist() if not cfg.MODEL.SEG.USE_SEG_POLY: masks = global_prediction.get_field("mask").cpu().numpy() else: masks = global_prediction.get_field("masks").get_polygons() result_logs = [] polygons = [] for k, box in enumerate(boxes): if box[2] - box[0] < 1 or box[3] - box[1] < 1: continue box = list(map(int, box)) if not cfg.MODEL.SEG.USE_SEG_POLY: mask = masks[k, 0, :, :] polygon = mask2polygon( mask, box, img.size, threshold=0.5, output_folder=output_folder ) else: polygon = list(masks[k].get_polygons()[0].cpu().numpy()) if not ("total_text" in output_folder or "cute80" in output_folder): polygon = polygon2rbox(polygon, height, width) if polygon is None: polygon = [ box[0], box[1], box[2], box[1], box[2], box[3], box[0], box[3], ] continue polygons.append(polygon) if cfg.MODEL.ROI_BOX_HEAD.INFERENCE_USE_BOX: score = scores[k] else: score = 1.0 if cfg.MODEL.CHAR_MASK_ON or cfg.SEQUENCE.SEQ_ON: if char_mask is None: word = 'aaa' rec_score = 1.0 char_score = None else: word = words[k] rec_score = rec_scores[k] char_score = rec_char_scoress[k] seq_word = seq_words[k] seq_char_scores = seq_scores[k] seq_score = sum(seq_char_scores) / float(len(seq_char_scores)) detailed_seq_score = detailed_seq_scores[k] detailed_seq_score = np.squeeze(np.array(detailed_seq_score), axis=1) else: word = 'aaa' rec_score = 1.0 char_score = [1.0, 1.0, 1.0] seq_word = 'aaa' seq_char_scores = [1.0, 1.0, 1.0] seq_score = 1.0 detailed_seq_score = None if "total_text" in output_folder or "cute80" in output_folder: result_log = ( [int(x * 1.0) for x in box[:4]] + polygon + [word] + [seq_word] + [score] + [rec_score] + [seq_score] + [char_score] + [detailed_seq_score] + [len(polygon)] ) else: result_log = ( [int(x * 1.0) for x in box[:4]] + polygon + [word] + [seq_word] + [score] + [rec_score] + [seq_score] + [char_score] + [detailed_seq_score] ) result_logs.append(result_log) if vis: colors = creat_color_map(37, 255) if cfg.MODEL.CHAR_MASK_ON: visualization(img, polygons, resize_ratio, colors, char_polygons, words) else: visualization(img, polygons, resize_ratio, colors) img.save(os.path.join(visu_dir, im_name)) format_output(results_dir, result_logs, im_name) def inference( model, data_loader, iou_types=("bbox",), box_only=False, device="cuda", expected_results=(), expected_results_sigma_tol=4, output_folder=None, model_name=None, cfg=None, ): # convert to a torch.device for efficiency model_name = model_name.split(".")[0] + "_" + str(cfg.INPUT.MIN_SIZE_TEST) predictions_path = os.path.join(output_folder, model_name + "_predictions.pth") seg_predictions_path = os.path.join( output_folder, model_name + "_seg_predictions.pth" ) # if os.path.isfile(predictions_path) and os.path.isfile(seg_predictions_path): if False: predictions = torch.load(predictions_path) seg_predictions = torch.load(seg_predictions_path) else: device = torch.device(device) num_devices = ( torch.distributed.get_world_size() if torch.distributed.is_initialized() else 1 ) logger = logging.getLogger("maskrcnn_benchmark.inference") dataset = data_loader.dataset logger.info("Start evaluation on {} images".format(len(dataset))) start_time = time.time() predictions, seg_predictions = compute_on_dataset( model, data_loader, device, cfg ) # wait for all processes to complete before measuring the time synchronize() total_time = time.time() - start_time total_time_str = str(datetime.timedelta(seconds=total_time)) logger.info( "Total inference time: {} ({} s / img per device, on {} devices)".format( total_time_str, total_time * num_devices / len(dataset), num_devices ) ) # predictions = _accumulate_predictions_from_multiple_gpus(predictions) # if not is_main_process(): # return if output_folder: torch.save(predictions, predictions_path) torch.save(seg_predictions, seg_predictions_path) prepare_results_for_evaluation( predictions, output_folder, model_name, seg_predictions=seg_predictions, vis=cfg.TEST.VIS, cfg=cfg )
23,471
40.839572
164
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/engine/trainer.py
#!/usr/bin/env python3 # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import datetime import logging import time import torch from maskrcnn_benchmark.utils.comm import get_world_size, is_main_process from maskrcnn_benchmark.utils.metric_logger import MetricLogger import torch.distributed as dist from apex import amp def reduce_loss_dict(loss_dict): """ Reduce the loss dictionary from all processes so that process with rank 0 has the averaged results. Returns a dict with the same fields as loss_dict, after reduction. """ world_size = get_world_size() if world_size < 2: return loss_dict with torch.no_grad(): loss_names = [] all_losses = [] for k, v in loss_dict.items(): loss_names.append(k) all_losses.append(v) all_losses = torch.stack(all_losses, dim=0) dist.reduce(all_losses, dst=0) if dist.get_rank() == 0: # only main process gets accumulated, so only divide by # world_size in this case all_losses /= world_size reduced_losses = {k: v for k, v in zip(loss_names, all_losses)} return reduced_losses def do_train( model, data_loader, optimizer, scheduler, checkpointer, device, checkpoint_period, arguments, tb_logger, cfg, local_rank, ): logger = logging.getLogger("maskrcnn_benchmark.trainer") logger.info("Start training") meters = MetricLogger(delimiter=" ") max_iter = len(data_loader) start_iter = arguments["iteration"] model.train() start_training_time = time.time() end = time.time() for iteration, (images, targets, _) in enumerate(data_loader, start_iter): data_time = time.time() - end arguments["iteration"] = iteration scheduler.step() images = images.to(device) targets = [target.to(device) for target in targets] loss_dict = model(images, targets) losses = sum(loss for loss in loss_dict.values()) # reduce losses over all GPUs for logging purposes loss_dict_reduced = reduce_loss_dict(loss_dict) losses_reduced = sum(loss for loss in loss_dict_reduced.values()) meters.update(loss=losses_reduced, **loss_dict_reduced) optimizer.zero_grad() # losses.backward() # Note: If mixed precision is not used, this ends up doing nothing # Otherwise apply loss scaling for mixed-precision recipe with amp.scale_loss(losses, optimizer) as scaled_losses: scaled_losses.backward() if cfg.SOLVER.USE_ADAM: torch.nn.utils.clip_grad_norm_(model.parameters(), 0.5) optimizer.step() batch_time = time.time() - end end = time.time() meters.update(time=batch_time, data=data_time) eta_seconds = meters.time.global_avg * (max_iter - iteration) eta_string = str(datetime.timedelta(seconds=int(eta_seconds))) if local_rank == 0 and (iteration % cfg.SOLVER.DISPLAY_FREQ == 0 or iteration == (max_iter - 1)): logger.info( meters.delimiter.join( [ "eta: {eta}", "iter: {iter}", "{meters}", "lr: {lr:.6f}", "max mem: {memory:.0f}", ] ).format( eta=eta_string, iter=iteration, meters=str(meters), lr=optimizer.param_groups[0]["lr"], memory=torch.cuda.max_memory_allocated() / 1024.0 / 1024.0, ) ) for tag, value in loss_dict_reduced.items(): tb_logger.scalar_summary(tag, value.item(), iteration) if local_rank == 0 and iteration % checkpoint_period == 0 and iteration > 0: checkpointer.save("model_{:07d}".format(iteration), **arguments) if local_rank == 0: checkpointer.save("model_{:07d}".format(iteration), **arguments) total_training_time = time.time() - start_training_time total_time_str = str(datetime.timedelta(seconds=total_training_time)) logger.info( "Total training time: {} ({:.4f} s / it)".format( total_time_str, total_training_time / (max_iter) ) )
4,397
34.184
105
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/c2_model_loading.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import logging import pickle from collections import OrderedDict import torch from maskrcnn_benchmark.utils.model_serialization import load_state_dict def _rename_basic_resnet_weights(layer_keys): layer_keys = [k.replace("_", ".") for k in layer_keys] layer_keys = [k.replace(".w", ".weight") for k in layer_keys] layer_keys = [k.replace(".bn", "_bn") for k in layer_keys] layer_keys = [k.replace(".b", ".bias") for k in layer_keys] layer_keys = [k.replace("_bn.s", "_bn.scale") for k in layer_keys] layer_keys = [k.replace(".biasranch", ".branch") for k in layer_keys] layer_keys = [k.replace("bbox.pred", "bbox_pred") for k in layer_keys] layer_keys = [k.replace("cls.score", "cls_score") for k in layer_keys] layer_keys = [k.replace("res.conv1_", "conv1_") for k in layer_keys] # RPN / Faster RCNN layer_keys = [k.replace(".biasbox", ".bbox") for k in layer_keys] layer_keys = [k.replace("conv.rpn", "rpn.conv") for k in layer_keys] layer_keys = [k.replace("rpn.bbox.pred", "rpn.bbox_pred") for k in layer_keys] layer_keys = [k.replace("rpn.cls.logits", "rpn.cls_logits") for k in layer_keys] # Affine-Channel -> BatchNorm enaming layer_keys = [k.replace("_bn.scale", "_bn.weight") for k in layer_keys] # Make torchvision-compatible layer_keys = [k.replace("conv1_bn.", "bn1.") for k in layer_keys] layer_keys = [k.replace("res2.", "layer1.") for k in layer_keys] layer_keys = [k.replace("res3.", "layer2.") for k in layer_keys] layer_keys = [k.replace("res4.", "layer3.") for k in layer_keys] layer_keys = [k.replace("res5.", "layer4.") for k in layer_keys] layer_keys = [k.replace(".branch2a.", ".conv1.") for k in layer_keys] layer_keys = [k.replace(".branch2a_bn.", ".bn1.") for k in layer_keys] layer_keys = [k.replace(".branch2b.", ".conv2.") for k in layer_keys] layer_keys = [k.replace(".branch2b_bn.", ".bn2.") for k in layer_keys] layer_keys = [k.replace(".branch2c.", ".conv3.") for k in layer_keys] layer_keys = [k.replace(".branch2c_bn.", ".bn3.") for k in layer_keys] layer_keys = [k.replace(".branch1.", ".downsample.0.") for k in layer_keys] layer_keys = [k.replace(".branch1_bn.", ".downsample.1.") for k in layer_keys] return layer_keys def _rename_fpn_weights(layer_keys, stage_names): for mapped_idx, stage_name in enumerate(stage_names, 1): suffix = "" if mapped_idx < 4: suffix = ".lateral" layer_keys = [ k.replace("fpn.inner.layer{}.sum{}".format(stage_name, suffix), "fpn_inner{}".format(mapped_idx)) for k in layer_keys ] layer_keys = [k.replace("fpn.layer{}.sum".format(stage_name), "fpn_layer{}".format(mapped_idx)) for k in layer_keys] layer_keys = [k.replace("rpn.conv.fpn2", "rpn.conv") for k in layer_keys] layer_keys = [k.replace("rpn.bbox_pred.fpn2", "rpn.bbox_pred") for k in layer_keys] layer_keys = [ k.replace("rpn.cls_logits.fpn2", "rpn.cls_logits") for k in layer_keys ] return layer_keys def _rename_weights_for_resnet(weights, stage_names): original_keys = sorted(weights.keys()) layer_keys = sorted(weights.keys()) # for X-101, rename output to fc1000 to avoid conflicts afterwards layer_keys = [k if k != "pred_b" else "fc1000_b" for k in layer_keys] layer_keys = [k if k != "pred_w" else "fc1000_w" for k in layer_keys] # performs basic renaming: _ -> . , etc layer_keys = _rename_basic_resnet_weights(layer_keys) # FPN layer_keys = _rename_fpn_weights(layer_keys, stage_names) # Mask R-CNN layer_keys = [k.replace("mask.fcn.logits", "mask_fcn_logits") for k in layer_keys] layer_keys = [k.replace(".[mask].fcn", "mask_fcn") for k in layer_keys] layer_keys = [k.replace("conv5.mask", "conv5_mask") for k in layer_keys] # Keypoint R-CNN layer_keys = [k.replace("kps.score.lowres", "kps_score_lowres") for k in layer_keys] layer_keys = [k.replace("kps.score", "kps_score") for k in layer_keys] layer_keys = [k.replace("conv.fcn", "conv_fcn") for k in layer_keys] # Rename for our RPN structure layer_keys = [k.replace("rpn.", "rpn.head.") for k in layer_keys] key_map = {k: v for k, v in zip(original_keys, layer_keys)} logger = logging.getLogger(__name__) logger.info("Remapping C2 weights") max_c2_key_size = max([len(k) for k in original_keys if "_momentum" not in k]) new_weights = OrderedDict() for k in original_keys: v = weights[k] if "_momentum" in k: continue # if 'fc1000' in k: # continue w = torch.from_numpy(v) # if "bn" in k: # w = w.view(1, -1, 1, 1) logger.info("C2 name: {: <{}} mapped name: {}".format(k, max_c2_key_size, key_map[k])) new_weights[key_map[k]] = w return new_weights def _load_c2_pickled_weights(file_path): with open(file_path, "rb") as f: data = pickle.load(f, encoding="latin1") if "blobs" in data: weights = data["blobs"] else: weights = data return weights def _rename_conv_weights_for_deformable_conv_layers(state_dict, cfg): import re logger = logging.getLogger(__name__) logger.info("Remapping conv weights for deformable conv weights") layer_keys = sorted(state_dict.keys()) for ix, stage_with_dcn in enumerate(cfg.MODEL.RESNETS.STAGE_WITH_DCN, 1): if not stage_with_dcn: continue for old_key in layer_keys: pattern = ".*layer{}.*conv2.*".format(ix) r = re.match(pattern, old_key) if r is None: continue for param in ["weight", "bias"]: if old_key.find(param) is -1: continue new_key = old_key.replace( "conv2.{}".format(param), "conv2.conv.{}".format(param) ) logger.info("pattern: {}, old_key: {}, new_key: {}".format( pattern, old_key, new_key )) state_dict[new_key] = state_dict[old_key] del state_dict[old_key] return state_dict _C2_STAGE_NAMES = { "R-50": ["1.2", "2.3", "3.5", "4.2"], "R-101": ["1.2", "2.3", "3.22", "4.2"], } def load_c2_format(cfg, f): # TODO make it support other architectures state_dict = _load_c2_pickled_weights(f) conv_body = cfg.MODEL.BACKBONE.CONV_BODY arch = conv_body.replace("-C4", "").replace("-FPN", "") stages = _C2_STAGE_NAMES[arch] state_dict = _rename_weights_for_resnet(state_dict, stages) # *********************************** # for deformable convolutional layer state_dict = _rename_conv_weights_for_deformable_conv_layers(state_dict, cfg) # *********************************** return dict(model=state_dict)
6,944
39.144509
129
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/metric_logger.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. from collections import defaultdict from collections import deque import torch class SmoothedValue(object): """Track a series of values and provide access to smoothed values over a window or the global series average. """ def __init__(self, window_size=20): self.deque = deque(maxlen=window_size) self.series = [] self.total = 0.0 self.count = 0 def update(self, value): self.deque.append(value) self.series.append(value) self.count += 1 self.total += value @property def median(self): d = torch.tensor(list(self.deque)) return d.median().item() @property def avg(self): d = torch.tensor(list(self.deque)) return d.mean().item() @property def global_avg(self): return self.total / self.count class MetricLogger(object): def __init__(self, delimiter="\t"): self.meters = defaultdict(SmoothedValue) self.delimiter = delimiter def update(self, **kwargs): for k, v in kwargs.items(): if isinstance(v, torch.Tensor): v = v.item() assert isinstance(v, (float, int)) self.meters[k].update(v) def __getattr__(self, attr): if attr in self.meters: return self.meters[attr] return object.__getattr__(self, attr) def __str__(self): loss_str = [] for name, meter in self.meters.items(): loss_str.append( "{}: {:.4f} ({:.4f})".format(name, meter.median, meter.global_avg) ) return self.delimiter.join(loss_str)
1,714
25.796875
82
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/checkpoint.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import logging import os import torch from maskrcnn_benchmark.utils.model_serialization import load_state_dict from maskrcnn_benchmark.utils.c2_model_loading import load_c2_format from maskrcnn_benchmark.utils.imports import import_file from maskrcnn_benchmark.utils.model_zoo import cache_url class Checkpointer(object): def __init__( self, model, optimizer=None, scheduler=None, save_dir="", save_to_disk=None, logger=None, ): self.model = model self.optimizer = optimizer self.scheduler = scheduler self.save_dir = save_dir self.save_to_disk = save_to_disk if logger is None: logger = logging.getLogger(__name__) self.logger = logger def save(self, name, **kwargs): if not self.save_dir: return if not self.save_to_disk: return data = {} data["model"] = self.model.state_dict() if self.optimizer is not None: data["optimizer"] = self.optimizer.state_dict() if self.scheduler is not None: data["scheduler"] = self.scheduler.state_dict() data.update(kwargs) save_file = os.path.join(self.save_dir, "{}.pth".format(name)) self.logger.info("Saving checkpoint to {}".format(save_file)) torch.save(data, save_file) self.tag_last_checkpoint(save_file) def load(self, f=None, resume=False): if self.has_checkpoint(): # override argument with existing checkpoint f = self.get_checkpoint_file() if not f: # no checkpoint could be found self.logger.info("No checkpoint found. Initializing model from scratch") return {} self.logger.info("Loading checkpoint from {}".format(f)) checkpoint = self._load_file(f) self._load_model(checkpoint) if resume: if "optimizer" in checkpoint and self.optimizer: self.logger.info("Loading optimizer from {}".format(f)) self.optimizer.load_state_dict(checkpoint.pop("optimizer")) if "scheduler" in checkpoint and self.scheduler: self.logger.info("Loading scheduler from {}".format(f)) self.scheduler.load_state_dict(checkpoint.pop("scheduler")) # return any further checkpoint data return checkpoint def has_checkpoint(self): save_file = os.path.join(self.save_dir, "last_checkpoint") return os.path.exists(save_file) def get_checkpoint_file(self): save_file = os.path.join(self.save_dir, "last_checkpoint") try: with open(save_file, "r") as f: last_saved = f.read() except IOError: # if file doesn't exist, maybe because it has just been # deleted by a separate process last_saved = "" return last_saved def tag_last_checkpoint(self, last_filename): save_file = os.path.join(self.save_dir, "last_checkpoint") with open(save_file, "w") as f: f.write(last_filename) def _load_file(self, f): return torch.load(f, map_location=torch.device("cpu")) def _load_model(self, checkpoint): load_state_dict(self.model, checkpoint.pop("model")) class DetectronCheckpointer(Checkpointer): def __init__( self, cfg, model, optimizer=None, scheduler=None, save_dir="", save_to_disk=None, logger=None, ): super(DetectronCheckpointer, self).__init__( model, optimizer, scheduler, save_dir, save_to_disk, logger ) self.cfg = cfg.clone() def _load_file(self, f): # catalog lookup if f.startswith("catalog://"): paths_catalog = import_file( "maskrcnn_benchmark.config.paths_catalog", self.cfg.PATHS_CATALOG, True ) catalog_f = paths_catalog.ModelCatalog.get(f[len("catalog://") :]) self.logger.info("{} points to {}".format(f, catalog_f)) f = catalog_f # download url files if f.startswith("http"): # if the file is a url path, download it and cache it cached_f = cache_url(f) self.logger.info("url {} cached in {}".format(f, cached_f)) f = cached_f # convert Caffe2 checkpoint from pkl if f.endswith(".pkl"): return load_c2_format(self.cfg, f) # load native detectron.pytorch checkpoint loaded = super(DetectronCheckpointer, self)._load_file(f) if "model" not in loaded: loaded = dict(model=loaded) return loaded
4,813
33.385714
87
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/comm.py
# # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. # """ # This file contains primitives for multi-gpu communication. # This is useful when doing distributed training. # """ # import os # import pickle # import tempfile # import time # import torch # import torch.distributed as dist # # def get_world_size(): # # if not dist.is_initialized(): # # return 1 # # return dist.get_world_size() # # # # # # def is_main_process(): # # if not dist.is_initialized(): # # return True # # return dist.get_rank() == 0 # # # # def get_rank(): # # if not dist.is_initialized(): # # return 0 # # return dist.get_rank() # # # # def synchronize(): # # """ # # Helper function to synchronize between multiple processes when # # using distributed training # # """ # # if not dist.is_initialized(): # # return # # world_size = dist.get_world_size() # # rank = dist.get_rank() # # if world_size == 1: # # return # # # # def _send_and_wait(r): # # if rank == r: # # tensor = torch.tensor(0, device="cuda") # # else: # # tensor = torch.tensor(1, device="cuda") # # dist.broadcast(tensor, r) # # while tensor.item() == 1: # # time.sleep(1) # # # # _send_and_wait(0) # # # now sync on the main process # # _send_and_wait(1) # # # # # def _encode(encoded_data, data): # # gets a byte representation for the data # encoded_bytes = pickle.dumps(data) # # convert this byte string into a byte tensor # storage = torch.ByteStorage.from_buffer(encoded_bytes) # tensor = torch.ByteTensor(storage).to("cuda") # # encoding: first byte is the size and then rest is the data # s = tensor.numel() # assert s <= 255, "Can't encode data greater than 255 bytes" # # put the encoded data in encoded_data # encoded_data[0] = s # encoded_data[1 : (s + 1)] = tensor # def _decode(encoded_data): # size = encoded_data[0] # encoded_tensor = encoded_data[1 : (size + 1)].to("cpu") # return pickle.loads(bytearray(encoded_tensor.tolist())) # # TODO try to use tensor in shared-memory instead of serializing to disk # # this involves getting the all_gather to work # def scatter_gather(data): # """ # This function gathers data from multiple processes, and returns them # in a list, as they were obtained from each process. # This function is useful for retrieving data from multiple processes, # when launching the code with torch.distributed.launch # Note: this function is slow and should not be used in tight loops, i.e., # do not use it in the training loop. # Arguments: # data: the object to be gathered from multiple processes. # It must be serializable # Returns: # result (list): a list with as many elements as there are processes, # where each element i in the list corresponds to the data that was # gathered from the process of rank i. # """ # # strategy: the main process creates a temporary directory, and communicates # # the location of the temporary directory to all other processes. # # each process will then serialize the data to the folder defined by # # the main process, and then the main process reads all of the serialized # # files and returns them in a list # if not dist.is_initialized(): # return [data] # synchronize() # # get rank of the current process # rank = dist.get_rank() # # the data to communicate should be small # data_to_communicate = torch.empty(256, dtype=torch.uint8, device="cuda") # if rank == 0: # # manually creates a temporary directory, that needs to be cleaned # # afterwards # tmp_dir = tempfile.mkdtemp() # _encode(data_to_communicate, tmp_dir) # synchronize() # # the main process (rank=0) communicates the data to all processes # dist.broadcast(data_to_communicate, 0) # # get the data that was communicated # tmp_dir = _decode(data_to_communicate) # # each process serializes to a different file # file_template = "file{}.pth" # tmp_file = os.path.join(tmp_dir, file_template.format(rank)) # torch.save(data, tmp_file) # # synchronize before loading the data # synchronize() # # only the master process returns the data # if rank == 0: # data_list = [] # world_size = dist.get_world_size() # for r in range(world_size): # file_path = os.path.join(tmp_dir, file_template.format(r)) # d = torch.load(file_path) # data_list.append(d) # # cleanup # os.remove(file_path) # # cleanup # os.rmdir(tmp_dir) # return data_list # def get_world_size(): # if not dist.is_available(): # print('distributed is not available') # return 1 # if not dist.is_initialized(): # print('distributed is not initialized') # return 1 # return dist.get_world_size() # def get_rank(): # if not dist.is_available(): # return 0 # if not dist.is_initialized(): # return 0 # return dist.get_rank() # def is_main_process(): # return get_rank() == 0 # def synchronize(): # """ # Helper function to synchronize (barrier) among all processes when # using distributed training # """ # if not dist.is_available(): # return # if not dist.is_initialized(): # return # world_size = dist.get_world_size() # if world_size == 1: # return # dist.barrier() # def all_gather(data): # """ # Run all_gather on arbitrary picklable data (not necessarily tensors) # Args: # data: any picklable object # Returns: # list[data]: list of data gathered from each rank # """ # world_size = get_world_size() # if world_size == 1: # return [data] # # serialized to a Tensor # buffer = pickle.dumps(data) # storage = torch.ByteStorage.from_buffer(buffer) # tensor = torch.ByteTensor(storage).to("cuda") # # obtain Tensor size of each rank # local_size = torch.IntTensor([tensor.numel()]).to("cuda") # size_list = [torch.IntTensor([0]).to("cuda") for _ in range(world_size)] # dist.all_gather(size_list, local_size) # size_list = [int(size.item()) for size in size_list] # max_size = max(size_list) # # receiving Tensor from all ranks # # we pad the tensor because torch all_gather does not support # # gathering tensors of different shapes # tensor_list = [] # for _ in size_list: # tensor_list.append(torch.ByteTensor(size=(max_size,)).to("cuda")) # if local_size != max_size: # padding = torch.ByteTensor(size=(max_size - local_size,)).to("cuda") # tensor = torch.cat((tensor, padding), dim=0) # dist.all_gather(tensor_list, tensor) # data_list = [] # for size, tensor in zip(size_list, tensor_list): # buffer = tensor.cpu().numpy().tobytes()[:size] # data_list.append(pickle.loads(buffer)) # return data_list # def reduce_dict(input_dict, average=True): # """ # Args: # input_dict (dict): all the values will be reduced # average (bool): whether to do average or sum # Reduce the values in the dictionary from all processes so that process with rank # 0 has the averaged results. Returns a dict with the same fields as # input_dict, after reduction. # """ # world_size = get_world_size() # if world_size < 2: # return input_dict # with torch.no_grad(): # names = [] # values = [] # # sort the keys so that they are consistent across processes # for k in sorted(input_dict.keys()): # names.append(k) # values.append(input_dict[k]) # values = torch.stack(values, dim=0) # dist.reduce(values, dst=0) # if dist.get_rank() == 0 and average: # # only main process gets accumulated, so only divide by # # world_size in this case # values /= world_size # reduced_dict = {k: v for k, v in zip(names, values)} # return reduced_dict """ This file contains primitives for multi-gpu communication. This is useful when doing distributed training. """ import pickle import time import torch import torch.distributed as dist def get_world_size(): if not dist.is_available(): return 1 if not dist.is_initialized(): return 1 return dist.get_world_size() def get_rank(): if not dist.is_available(): return 0 if not dist.is_initialized(): return 0 return dist.get_rank() def is_main_process(): return get_rank() == 0 def synchronize(): """ Helper function to synchronize (barrier) among all processes when using distributed training """ if not dist.is_available(): return if not dist.is_initialized(): return world_size = dist.get_world_size() if world_size == 1: return dist.barrier() def scatter_gather(data): """ Run all_gather on arbitrary picklable data (not necessarily tensors) Args: data: any picklable object Returns: list[data]: list of data gathered from each rank """ world_size = get_world_size() if world_size == 1: return [data] # serialized to a Tensor buffer = pickle.dumps(data) storage = torch.ByteStorage.from_buffer(buffer) tensor = torch.ByteTensor(storage).to("cuda") # obtain Tensor size of each rank local_size = torch.LongTensor([tensor.numel()]).to("cuda") size_list = [torch.LongTensor([0]).to("cuda") for _ in range(world_size)] dist.all_gather(size_list, local_size) size_list = [int(size.item()) for size in size_list] max_size = max(size_list) # receiving Tensor from all ranks # we pad the tensor because torch all_gather does not support # gathering tensors of different shapes tensor_list = [] for _ in size_list: tensor_list.append(torch.ByteTensor(size=(max_size,)).to("cuda")) if local_size != max_size: padding = torch.ByteTensor(size=(max_size - local_size,)).to("cuda") tensor = torch.cat((tensor, padding), dim=0) dist.all_gather(tensor_list, tensor) data_list = [] for size, tensor in zip(size_list, tensor_list): buffer = tensor.cpu().numpy().tobytes()[:size] data_list.append(pickle.loads(buffer)) return data_list def reduce_dict(input_dict, average=True): """ Args: input_dict (dict): all the values will be reduced average (bool): whether to do average or sum Reduce the values in the dictionary from all processes so that process with rank 0 has the averaged results. Returns a dict with the same fields as input_dict, after reduction. """ world_size = get_world_size() if world_size < 2: return input_dict with torch.no_grad(): names = [] values = [] # sort the keys so that they are consistent across processes for k in sorted(input_dict.keys()): names.append(k) values.append(input_dict[k]) values = torch.stack(values, dim=0) dist.reduce(values, dst=0) if dist.get_rank() == 0 and average: # only main process gets accumulated, so only divide by # world_size in this case values /= world_size reduced_dict = {k: v for k, v in zip(names, values)} return reduced_dict
11,687
29.83905
86
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/registry.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. def _register_generic(module_dict, module_name, module): assert module_name not in module_dict module_dict[module_name] = module class Registry(dict): ''' A helper class for managing registering modules, it extends a dictionary and provides a register functions. Eg. creeting a registry: some_registry = Registry({"default": default_module}) There're two ways of registering new modules: 1): normal way is just calling register function: def foo(): ... some_registry.register("foo_module", foo) 2): used as decorator when declaring the module: @some_registry.register("foo_module") @some_registry.register("foo_modeul_nickname") def foo(): ... Access of module is just like using a dictionary, eg: f = some_registry["foo_modeul"] ''' def __init__(self, *args, **kwargs): super(Registry, self).__init__(*args, **kwargs) def register(self, module_name, module=None): # used as function call if module is not None: _register_generic(self, module_name, module) return # used as decorator def register_fn(fn): _register_generic(self, module_name, fn) return fn return register_fn
1,384
29.777778
76
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/model_zoo.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import os import sys try: from torch.hub import _download_url_to_file from torch.hub import urlparse from torch.hub import HASH_REGEX except ImportError: from torch.utils.model_zoo import _download_url_to_file from torch.utils.model_zoo import urlparse from torch.utils.model_zoo import HASH_REGEX from maskrcnn_benchmark.utils.comm import is_main_process from maskrcnn_benchmark.utils.comm import synchronize # very similar to https://github.com/pytorch/pytorch/blob/master/torch/utils/model_zoo.py # but with a few improvements and modifications def cache_url(url, model_dir=None, progress=True): r"""Loads the Torch serialized object at the given URL. If the object is already present in `model_dir`, it's deserialized and returned. The filename part of the URL should follow the naming convention ``filename-<sha256>.ext`` where ``<sha256>`` is the first eight or more digits of the SHA256 hash of the contents of the file. The hash is used to ensure unique names and to verify the contents of the file. The default value of `model_dir` is ``$TORCH_HOME/models`` where ``$TORCH_HOME`` defaults to ``~/.torch``. The default directory can be overridden with the ``$TORCH_MODEL_ZOO`` environment variable. Args: url (string): URL of the object to download model_dir (string, optional): directory in which to save the object progress (bool, optional): whether or not to display a progress bar to stderr Example: >>> cached_file = maskrcnn_benchmark.utils.model_zoo.cache_url('https://s3.amazonaws.com/pytorch/models/resnet18-5c106cde.pth') """ if model_dir is None: torch_home = os.path.expanduser(os.getenv("TORCH_HOME", "~/.torch")) model_dir = os.getenv("TORCH_MODEL_ZOO", os.path.join(torch_home, "models")) if not os.path.exists(model_dir): os.makedirs(model_dir) parts = urlparse(url) filename = os.path.basename(parts.path) if filename == "model_final.pkl": # workaround as pre-trained Caffe2 models from Detectron have all the same filename # so make the full path the filename by replacing / with _ filename = parts.path.replace("/", "_") cached_file = os.path.join(model_dir, filename) if not os.path.exists(cached_file) and is_main_process(): sys.stderr.write('Downloading: "{}" to {}\n'.format(url, cached_file)) hash_prefix = HASH_REGEX.search(filename) if hash_prefix is not None: hash_prefix = hash_prefix.group(1) # workaround: Caffe2 models don't have a hash, but follow the R-50 convention, # which matches the hash PyTorch uses. So we skip the hash matching # if the hash_prefix is less than 6 characters if len(hash_prefix) < 6: hash_prefix = None _download_url_to_file(url, cached_file, hash_prefix, progress=progress) synchronize() return cached_file
3,044
48.918033
135
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/logging.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import logging import os import sys from tensorboardX import SummaryWriter def setup_logger(name, save_dir, distributed_rank=0): logger = logging.getLogger(name) logger.setLevel(logging.DEBUG) # don't log results for the non-master process if distributed_rank > 0: return logger ch = logging.StreamHandler(stream=sys.stdout) ch.setLevel(logging.DEBUG) formatter = logging.Formatter("%(asctime)s %(name)s %(levelname)s: %(message)s") ch.setFormatter(formatter) logger.addHandler(ch) if save_dir: fh = logging.FileHandler(os.path.join(save_dir, "log.txt")) fh.setLevel(logging.DEBUG) fh.setFormatter(formatter) logger.addHandler(fh) return logger class Logger(object): def __init__(self, log_dir, distributed_rank=0): """Create a summary writer logging to log_dir.""" self.distributed_rank = distributed_rank if distributed_rank == 0: self.writer = SummaryWriter(log_dir) def scalar_summary(self, tag, value, step): """Log a scalar variable.""" if self.distributed_rank == 0: self.writer.add_scalar(tag, value, step)
1,252
28.833333
84
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/collect_env.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import PIL from torch.utils.collect_env import get_pretty_env_info def get_pil_version(): return "\n Pillow ({})".format(PIL.__version__) def collect_env_info(): env_str = get_pretty_env_info() env_str += get_pil_version() return env_str
338
21.6
71
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/model_serialization.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. from collections import OrderedDict import logging import torch from maskrcnn_benchmark.utils.imports import import_file def align_and_update_state_dicts(model_state_dict, loaded_state_dict): """ Strategy: suppose that the models that we will create will have prefixes appended to each of its keys, for example due to an extra level of nesting that the original pre-trained weights from ImageNet won't contain. For example, model.state_dict() might return backbone[0].body.res2.conv1.weight, while the pre-trained model contains res2.conv1.weight. We thus want to match both parameters together. For that, we look for each model weight, look among all loaded keys if there is one that is a suffix of the current weight name, and use it if that's the case. If multiple matches exist, take the one with longest size of the corresponding name. For example, for the same model as before, the pretrained weight file can contain both res2.conv1.weight, as well as conv1.weight. In this case, we want to match backbone[0].body.conv1.weight to conv1.weight, and backbone[0].body.res2.conv1.weight to res2.conv1.weight. """ current_keys = sorted(list(model_state_dict.keys())) loaded_keys = sorted(list(loaded_state_dict.keys())) # get a matrix of string matches, where each (i, j) entry correspond to the size of the # loaded_key string, if it matches match_matrix = [ len(j) if i.endswith(j) else 0 for i in current_keys for j in loaded_keys ] match_matrix = torch.as_tensor(match_matrix).view( len(current_keys), len(loaded_keys) ) max_match_size, idxs = match_matrix.max(1) # remove indices that correspond to no-match idxs[max_match_size == 0] = -1 # used for logging max_size = max([len(key) for key in current_keys]) if current_keys else 1 max_size_loaded = max([len(key) for key in loaded_keys]) if loaded_keys else 1 log_str_template = "{: <{}} loaded from {: <{}} of shape {}" logger = logging.getLogger(__name__) for idx_new, idx_old in enumerate(idxs.tolist()): if idx_old == -1: continue key = current_keys[idx_new] key_old = loaded_keys[idx_old] model_state_dict[key] = loaded_state_dict[key_old] logger.info( log_str_template.format( key, max_size, key_old, max_size_loaded, tuple(loaded_state_dict[key_old].shape), ) ) def strip_prefix_if_present(state_dict, prefix): keys = sorted(state_dict.keys()) if not all(key.startswith(prefix) for key in keys): return state_dict stripped_state_dict = OrderedDict() for key, value in state_dict.items(): stripped_state_dict[key.replace(prefix, "")] = value return stripped_state_dict def load_state_dict(model, loaded_state_dict): model_state_dict = model.state_dict() # if the state_dict comes from a model that was wrapped in a # DataParallel or DistributedDataParallel during serialization, # remove the "module" prefix before performing the matching loaded_state_dict = strip_prefix_if_present(loaded_state_dict, prefix="module.") align_and_update_state_dicts(model_state_dict, loaded_state_dict) # use strict loading model.load_state_dict(model_state_dict)
3,464
41.777778
91
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/chars.py
import os import cv2 import numpy as np def char2num(char): if char in "0123456789": num = ord(char) - ord("0") + 1 elif char in "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ": num = ord(char.lower()) - ord("a") + 11 else: num = 0 return num def num2char(num): chars = "_0123456789abcdefghijklmnopqrstuvwxyz" char = chars[num] # if num >=1 and num <=10: # char = chr(ord('0') + num - 1) # elif num > 10 and num <= 36: # char = chr(ord('a') + num - 11) # else: # print('error number:%d'%(num)) # exit() return char def getstr_grid(seg, box, threshold=192): pos = 255 - (seg[0] * 255).astype(np.uint8) mask_index = np.argmax(seg, axis=0) mask_index = mask_index.astype(np.uint8) pos = pos.astype(np.uint8) string, score, rec_scores, char_polygons = seg2text( pos, mask_index, seg, box, threshold=threshold ) return string, score, rec_scores, char_polygons def seg2text(gray, mask, seg, box, threshold=192): ## input numpy img_h, img_w = gray.shape box_w = box[2] - box[0] box_h = box[3] - box[1] ratio_h = float(box_h) / img_h ratio_w = float(box_w) / img_w # SE1=cv2.getStructuringElement(cv2.MORPH_RECT,(3,3)) # gray = cv2.erode(gray,SE1) # gray = cv2.dilate(gray,SE1) # gray = cv2.morphologyEx(gray,cv2.MORPH_CLOSE,SE1) ret, thresh = cv2.threshold(gray, threshold, 255, cv2.THRESH_BINARY) try: _, contours, _ = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE) except: contours, _ = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE) chars = [] scores = [] char_polygons = [] for i in range(len(contours)): char = {} temp = np.zeros((img_h, img_w)).astype(np.uint8) cv2.drawContours(temp, [contours[i]], 0, (255), -1) x, y, w, h = cv2.boundingRect(contours[i]) c_x, c_y = x + w / 2, y + h / 2 perimeter = cv2.arcLength(contours[i], True) epsilon = 0.01 * cv2.arcLength(contours[i], True) approx = cv2.approxPolyDP(contours[i], epsilon, True) pts = approx.reshape((-1, 2)) pts[:, 0] = pts[:, 0] * ratio_w + box[0] pts[:, 1] = pts[:, 1] * ratio_h + box[1] polygon = list(pts.reshape((-1,))) polygon = list(map(int, polygon)) if len(polygon) >= 6: char_polygons.append(polygon) # x1 = x * ratio_w + box[0] # y1 = y * ratio_h + box[1] # x3 = (x + w) * ratio_w + box[0] # y3 = (y + h) * ratio_h + box[1] # polygon = [x1, y1, x3, y1, x3, y3, x1, y3] regions = seg[1:, temp == 255].reshape((36, -1)) cs = np.mean(regions, axis=1) sym = num2char(np.argmax(cs.reshape((-1))) + 1) char["x"] = c_x char["y"] = c_y char["s"] = sym char["cs"] = cs.reshape((-1, 1)) scores.append(np.max(char["cs"], axis=0)[0]) chars.append(char) chars = sorted(chars, key=lambda x: x["x"]) string = "" css = [] for char in chars: string = string + char["s"] css.append(char["cs"]) if len(scores) > 0: score = sum(scores) / len(scores) else: score = 0.00 if not css: css = [0.0] return string, score, np.hstack(css), char_polygons # def get_tight_rect(points, start_x, start_y, image_height, image_width, scale): # points = list(points) # ps = sorted(points, key=lambda x: x[0]) # # if ps[1][1] > ps[0][1]: # px1 = ps[0][0] * scale + start_x # py1 = ps[0][1] * scale + start_y # px4 = ps[1][0] * scale + start_x # py4 = ps[1][1] * scale + start_y # else: # px1 = ps[1][0] * scale + start_x # py1 = ps[1][1] * scale + start_y # px4 = ps[0][0] * scale + start_x # py4 = ps[0][1] * scale + start_y # if ps[3][1] > ps[2][1]: # px2 = ps[2][0] * scale + start_x # py2 = ps[2][1] * scale + start_y # px3 = ps[3][0] * scale + start_x # py3 = ps[3][1] * scale + start_y # else: # px2 = ps[3][0] * scale + start_x # py2 = ps[3][1] * scale + start_y # px3 = ps[2][0] * scale + start_x # py3 = ps[2][1] * scale + start_y # # if px1 < 0: # px1 = 1 # if px1 > image_width: # px1 = image_width - 1 # if px2 < 0: # px2 = 1 # if px2 > image_width: # px2 = image_width - 1 # if px3 < 0: # px3 = 1 # if px3 > image_width: # px3 = image_width - 1 # if px4 < 0: # px4 = 1 # if px4 > image_width: # px4 = image_width - 1 # # if py1 < 0: # py1 = 1 # if py1 > image_height: # py1 = image_height - 1 # if py2 < 0: # py2 = 1 # if py2 > image_height: # py2 = image_height - 1 # if py3 < 0: # py3 = 1 # if py3 > image_height: # py3 = image_height - 1 # if py4 < 0: # py4 = 1 # if py4 > image_height: # py4 = image_height - 1 # return [px1, py1, px2, py2, px3, py3, px4, py4] def get_tight_rect(points, start_x, start_y, image_height, image_width, scale): points = list(points) ps = sorted(points, key=lambda x: x[0]) if ps[1][1] > ps[0][1]: px1 = ps[0][0] * scale + start_x py1 = ps[0][1] * scale + start_y px4 = ps[1][0] * scale + start_x py4 = ps[1][1] * scale + start_y else: px1 = ps[1][0] * scale + start_x py1 = ps[1][1] * scale + start_y px4 = ps[0][0] * scale + start_x py4 = ps[0][1] * scale + start_y if ps[3][1] > ps[2][1]: px2 = ps[2][0] * scale + start_x py2 = ps[2][1] * scale + start_y px3 = ps[3][0] * scale + start_x py3 = ps[3][1] * scale + start_y else: px2 = ps[3][0] * scale + start_x py2 = ps[3][1] * scale + start_y px3 = ps[2][0] * scale + start_x py3 = ps[2][1] * scale + start_y px1 = min(max(px1, 1), image_width - 1) px2 = min(max(px2, 1), image_width - 1) px3 = min(max(px3, 1), image_width - 1) px4 = min(max(px4, 1), image_width - 1) py1 = min(max(py1, 1), image_height - 1) py2 = min(max(py2, 1), image_height - 1) py3 = min(max(py3, 1), image_height - 1) py4 = min(max(py4, 1), image_height - 1) return [px1, py1, px2, py2, px3, py3, px4, py4]
6,419
31.1
89
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/__init__.py
0
0
0
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/miscellaneous.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import errno import os def mkdir(path): try: os.makedirs(path) except OSError as e: if e.errno != errno.EEXIST: raise
228
18.083333
71
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/env.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import os from maskrcnn_benchmark.utils.imports import import_file def setup_environment(): """Perform environment setup work. The default setup is a no-op, but this function allows the user to specify a Python source file that performs custom setup work that may be necessary to their computing environment. """ custom_module_path = os.environ.get("TORCH_DETECTRON_ENV_MODULE") if custom_module_path: setup_custom_environment(custom_module_path) else: # The default setup is a no-op pass def setup_custom_environment(custom_module_path): """Load custom environment setup from a Python source file and run the setup function. """ module = import_file("maskrcnn_benchmark.utils.env.custom_module", custom_module_path) assert hasattr(module, "setup_environment") and callable( module.setup_environment ), ( "Custom environment module defined in {} does not have the " "required callable attribute 'setup_environment'." ).format( custom_module_path ) module.setup_environment() # Force environment setup when this module is imported setup_environment()
1,249
31.894737
90
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/utils/imports.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import importlib import importlib.util import sys # from https://stackoverflow.com/questions/67631/how-to-import-a-module-given-the-full-path?utm_medium=organic&utm_source=google_rich_qa&utm_campaign=google_rich_qa def import_file(module_name, file_path, make_importable=False): spec = importlib.util.spec_from_file_location(module_name, file_path) module = importlib.util.module_from_spec(spec) spec.loader.exec_module(module) if make_importable: sys.modules[module_name] = module return module
598
38.933333
164
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/data/__init__.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. from .build import make_data_loader
108
35.333333
71
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/data/collate_batch.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. from maskrcnn_benchmark.structures.image_list import to_image_list, to_image_target_list class BatchCollator(object): """ From a list of samples from the dataset, returns the batched images and targets. This should be passed to the DataLoader """ def __init__(self, size_divisible=0): self.size_divisible = size_divisible def __call__(self, batch): transposed_batch = list(zip(*batch)) images = to_image_list(transposed_batch[0], self.size_divisible) targets = transposed_batch[1] img_ids = transposed_batch[2] # if transposed_batch[1] is None: # images = to_image_list(transposed_batch[0], self.size_divisible) # targets = transposed_batch[1] # img_ids = transposed_batch[2] # else: # images, targets = to_image_target_list(transposed_batch[0], self.size_divisible, transposed_batch[1]) # img_ids = transposed_batch[2] return images, targets, img_ids
1,080
37.607143
115
py
MaskTextSpotterV3
MaskTextSpotterV3-master/maskrcnn_benchmark/data/build.py
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. import bisect import logging import torch.utils.data from maskrcnn_benchmark.utils.comm import get_world_size from maskrcnn_benchmark.utils.imports import import_file from . import datasets as D from . import samplers from .collate_batch import BatchCollator from .transforms import build_transforms def build_dataset(cfg, dataset_list, transforms, dataset_catalog, is_train=True): """ Arguments: dataset_list (list[str]): Contains the names of the datasets, i.e., coco_2014_trian, coco_2014_val, etc transforms (callable): transforms to apply to each (image, target) sample dataset_catalog (DatasetCatalog): contains the information on how to construct a dataset. is_train (bool): whether to setup the dataset for training or testing """ if not isinstance(dataset_list, (list, tuple)): raise RuntimeError( "dataset_list should be a list of strings, got {}".format(dataset_list)) datasets = [] for dataset_name in dataset_list: data = dataset_catalog.get(dataset_name) factory = getattr(D, data["factory"]) args = data["args"] # for COCODataset, we want to remove images without annotations # during training if data["factory"] == "COCODataset": args["remove_images_without_annotations"] = is_train args["transforms"] = transforms args["ignore_difficult"] = cfg.DATASETS.IGNORE_DIFFICULT # make dataset from factory dataset = factory(**args) datasets.append(dataset) # for testing, return a list of datasets if not is_train: return datasets # for training, concatenate all datasets into a single one dataset = datasets[0] if len(datasets) > 1: dataset = D.MixDataset(datasets, cfg.DATASETS.RATIOS) # dataset = D.ConcatDataset(datasets) return [dataset] def make_data_sampler(dataset, shuffle, distributed): if distributed: return samplers.DistributedSampler(dataset, shuffle=shuffle) if shuffle: sampler = torch.utils.data.sampler.RandomSampler(dataset) else: sampler = torch.utils.data.sampler.SequentialSampler(dataset) return sampler def _quantize(x, bins): bins = sorted(bins.copy()) quantized = list(map(lambda y: bisect.bisect_right(bins, y), x)) return quantized def _compute_aspect_ratios(dataset): aspect_ratios = [] for i in range(len(dataset)): img_info = dataset.get_img_info(i) aspect_ratio = float(img_info["height"]) / float(img_info["width"]) aspect_ratios.append(aspect_ratio) return aspect_ratios def make_batch_data_sampler( dataset, sampler, aspect_grouping, images_per_batch, num_iters=None, start_iter=0 ): if aspect_grouping: if not isinstance(aspect_grouping, (list, tuple)): aspect_grouping = [aspect_grouping] aspect_ratios = _compute_aspect_ratios(dataset) group_ids = _quantize(aspect_ratios, aspect_grouping) batch_sampler = samplers.GroupedBatchSampler( sampler, group_ids, images_per_batch, drop_uneven=False ) else: batch_sampler = torch.utils.data.sampler.BatchSampler( sampler, images_per_batch, drop_last=False ) if num_iters is not None: batch_sampler = samplers.IterationBasedBatchSampler(batch_sampler, num_iters, start_iter) return batch_sampler def make_data_loader(cfg, is_train=True, is_distributed=False, start_iter=0): num_gpus = get_world_size() if is_train: images_per_batch = cfg.SOLVER.IMS_PER_BATCH assert ( images_per_batch % num_gpus == 0 ), "SOLVER.IMS_PER_BATCH ({}) must be divisible by the number " "of GPUs ({}) used.".format(images_per_batch, num_gpus) images_per_gpu = images_per_batch // num_gpus shuffle = True num_iters = cfg.SOLVER.MAX_ITER else: images_per_batch = cfg.TEST.IMS_PER_BATCH assert ( images_per_batch % num_gpus == 0 ), "TEST.IMS_PER_BATCH ({}) must be divisible by the number " "of GPUs ({}) used.".format(images_per_batch, num_gpus) images_per_gpu = images_per_batch // num_gpus shuffle = False if not is_distributed else True num_iters = None start_iter = 0 if images_per_gpu > 1: logger = logging.getLogger(__name__) logger.warning( "When using more than one image per GPU you may encounter " "an out-of-memory (OOM) error if your GPU does not have " "sufficient memory. If this happens, you can reduce " "SOLVER.IMS_PER_BATCH (for training) or " "TEST.IMS_PER_BATCH (for inference). For training, you must " "also adjust the learning rate and schedule length according " "to the linear scaling rule. See for example: " "https://github.com/facebookresearch/Detectron/blob/master/configs/getting_started/tutorial_1gpu_e2e_faster_rcnn_R-50-FPN.yaml#L14" ) # group images which have similar aspect ratio. In this case, we only # group in two cases: those with width / height > 1, and the other way around, # but the code supports more general grouping strategy aspect_grouping = [1] if cfg.DATALOADER.ASPECT_RATIO_GROUPING else [] paths_catalog = import_file( "maskrcnn_benchmark.config.paths_catalog", cfg.PATHS_CATALOG, True ) DatasetCatalog = paths_catalog.DatasetCatalog dataset_list = cfg.DATASETS.TRAIN if is_train else cfg.DATASETS.TEST transforms = build_transforms(cfg, is_train) datasets = build_dataset(cfg,dataset_list, transforms, DatasetCatalog, is_train) data_loaders = [] for dataset in datasets: ''' for i in range(20): a=dataset[i] ipdb.set_trace() ''' sampler = make_data_sampler(dataset, shuffle, is_distributed) batch_sampler = make_batch_data_sampler( dataset, sampler, aspect_grouping, images_per_gpu, num_iters, start_iter ) collator = BatchCollator(cfg.DATALOADER.SIZE_DIVISIBILITY) num_workers = cfg.DATALOADER.NUM_WORKERS data_loader = torch.utils.data.DataLoader( dataset, num_workers=num_workers, batch_sampler=batch_sampler, collate_fn=collator, ) data_loaders.append(data_loader) if is_train: # during training, a single (possibly concatenated) data_loader is returned assert len(data_loaders) == 1 return data_loaders[0] return data_loaders
6,740
37.301136
143
py