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
libai
libai-main/projects/QQP/modeling/load_megatron_weight.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import logging import oneflow as flow import torch import libai.utils.distributed as dist from libai.utils.checkpoint import get_missing_parameters_message, get_unexpected_parameters_message logger = logging.getLogger("libai." + __name__) def convert_tensor(tensor: torch.Tensor): tensor = tensor.float() return flow.Tensor(tensor.cpu().numpy()) def change_megatron_key(state_dict): of_state_dict = {} # Language model. language_model = state_dict["language_model"] # Embedding. embedding = language_model["embedding"] of_state_dict["embeddings.vocab_embeddings.weight"] = convert_tensor( embedding["word_embeddings"]["weight"] ) of_state_dict["embeddings.position_embeddings.weight"] = convert_tensor( embedding["position_embeddings"]["weight"] ) of_state_dict["embeddings.tokentype_embeddings.weight"] = convert_tensor( embedding["tokentype_embeddings"]["weight"] ) # Encoder. encoder = language_model["encoder"] for key, value in encoder.items(): # Change layers.0.input_layernorm.weight -> encoder.layers_0.input_layernorm.weight key = "encoders." + key.replace("layers.", "") if key.startswith("encoders.final_layernorm"): key = key.replace("encoders.", "") of_state_dict[key] = convert_tensor(value) # Pooler. pooler = language_model["pooler"] of_state_dict["pooler.dense.weight"] = convert_tensor(pooler["dense.weight"]) of_state_dict["pooler.dense.bias"] = convert_tensor(pooler["dense.bias"]) # LM head. lm_head = state_dict["lm_head"] of_state_dict["cls.predictions.dense.weight"] = convert_tensor(lm_head["dense.weight"]) of_state_dict["cls.predictions.dense.bias"] = convert_tensor(lm_head["dense.bias"]) of_state_dict["cls.predictions.layernorm.weight"] = convert_tensor(lm_head["layernorm.weight"]) of_state_dict["cls.predictions.layernorm.bias"] = convert_tensor(lm_head["layernorm.bias"]) of_state_dict["lm_logits.bias"] = convert_tensor(lm_head["bias"]) # Binary head. binary_head = state_dict["binary_head"] of_state_dict["cls.seq_relationship.weight"] = convert_tensor(binary_head["weight"]) of_state_dict["cls.seq_relationship.bias"] = convert_tensor((binary_head["bias"])) return of_state_dict def load_tensor(tensor_lhs, tensor_rhs): tensor_rhs = flow.to_global( tensor_rhs, placement=tensor_lhs.placement, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), ) tensor_rhs = tensor_rhs.to_global(sbp=tensor_lhs.sbp) tensor_lhs.copy_(tensor_rhs) def load_model(model: flow.nn.Module, state_dict): model_state_dict = model.state_dict() # Decide shape incorrect_shapes = [] for k in list(state_dict.keys()): if k in model_state_dict: shape_model = tuple(model_state_dict[k].shape) shape_ckpt = tuple(state_dict[k].shape) if shape_model != shape_ckpt: incorrect_shapes.append((k, shape_ckpt, shape_model)) state_dict.pop(k) unexpected_keys = [] for key, value in state_dict.items(): if key not in model_state_dict: unexpected_keys.append(key) continue model_state_dict.pop(key) load_tensor(model.state_dict()[key], value) missing_keys = list(model_state_dict.keys()) for k, shape_checkpoint, shape_model in incorrect_shapes: logger.warning( "Skip loading parameter '{}' to the model due to incompatible " "shapes: {} in the checkpoint but {} in the " "model! You might want to double check if this is expected.".format( k, shape_checkpoint, shape_model ) ) if missing_keys: logger.info(get_missing_parameters_message(missing_keys)) if unexpected_keys: logger.info(get_unexpected_parameters_message(unexpected_keys)) def load_megatron_bert(model: flow.nn.Module, model_weight_path: str): import torch megatron_state_dict = torch.load(model_weight_path, map_location="cpu")["model"] of_state_dict = change_megatron_key(megatron_state_dict) load_model(model, of_state_dict)
4,859
35
100
py
libai
libai-main/projects/DALLE2/dalle2_inference.py
import os from typing import Dict import oneflow as flow from dalle2.dalle2_loader import Dalle2ModelLoader from dalle2.model_weights.download_utils import download_dalle2_weights from dalle2.tokenizer import SimpleTokenizer from oneflow.framework import balanced_splitter import libai.utils.distributed as dist from libai.inference.basic import BasePipeline class Dalle2Pipeline(BasePipeline): def __init__( self, config_file, data_parallel=None, tensor_parallel=None, pipeline_parallel=None, pipeline_stage_id=None, pipeline_num_layers=None, model_path=None, mode="libai", **kwargs, ): super().__init__( config_file, data_parallel, tensor_parallel, pipeline_parallel, pipeline_stage_id, model_path, pipeline_num_layers, mode, **kwargs, ) def update_cfg( self, data_parallel=1, tensor_parallel=1, pipeline_parallel=1, pipeline_stage_id=None, pipeline_num_layers=None, ): super().update_cfg( data_parallel, tensor_parallel, pipeline_parallel, pipeline_stage_id, pipeline_num_layers, ) self.cfg.model.prior.clip.name = "./dalle2/model_weights/ViT-L-14.pt" self.cfg.model.prior_weight_path = "./dalle2/model_weights/prior_aes_finetune.pth" self.cfg.model.decoder_weight_path = "./dalle2/model_weights/latest.pth" self.cfg.swinir.swinir_path = ( "./swinir/weights/003_realSR_BSRGAN_DFOWMFC_s64w8_SwinIR-L_x4_GAN.pth" ) def load_pretrain_weight(self, libai_cfg_model, model_path, mode=None): if dist.is_main_process(): download_dalle2_weights(self.cfg) dist.synchronize() model_loader = Dalle2ModelLoader(libai_cfg_model, self.cfg, model_path) return model_loader.load() def build_tokenizer(self, cfg): return SimpleTokenizer() # return instantiate(cfg.tokenizer) def _parse_parameters(self, model_path=None, save_images=False, upsample_scale=None, **kwargs): preprocess_params = {} forward_params = { "model_path": model_path, "num_samples_per_batch": kwargs.get("num_samples_per_batch", 2), "prior_cond_scale": kwargs.get("prior_cond_scale", 1.0), "decoder_cond_scale": kwargs.get("decoder_cond_scale", 3.5), } postprocess_params = { "save_images": save_images, "upsample_scale": upsample_scale, "swinir_path": "./swinir/weights/003_realSR_BSRGAN_DFOWMFC_s64w8_SwinIR-L_x4_GAN.pth", } return preprocess_params, forward_params, postprocess_params def split_data(self, text): rank = dist.get_rank() indices = balanced_splitter.BalancedRanges(len(text), dist.get_world_size()) return text[indices[rank][0] : indices[rank][1]] def preprocess(self, input_, **preprocess_parameters: Dict) -> dict: tokens = self.tokenizer.tokenize(input_).to_global( placement=flow.placement(type="cuda", ranks=list(range(dist.get_world_size()))), sbp=flow.sbp.broadcast, ) return {"text": input_, "tokens": tokens} def forward(self, model_input_dict, **forward_params) -> dict: tokens = model_input_dict["tokens"] text_embed, text_encodings, text_mask = self.model.prior.clip.embed_text(tokens) image_embed = self.model.prior.sample( tokens, num_samples_per_batch=forward_params["num_samples_per_batch"], cond_scale=forward_params["prior_cond_scale"], ) image_embed = self.model.decoder.sample( image_embed=image_embed, text_encodings=text_encodings, text_mask=text_mask, cond_scale=forward_params["decoder_cond_scale"], ) return {"image_embed": image_embed} def postprocess(self, model_output_dict, **postprocess_params: Dict) -> dict: if not postprocess_params.get("save_images", False): return model_output_dict output_path = postprocess_params.get("output_dit", "./outputs") os.makedirs(output_path, exist_ok=True) import flowvision.transforms as T to_pil = T.ToPILImage() images = model_output_dict["image_embed"].to("cpu") images_64x64 = list(map(to_pil, [images[i] for i in range(images.shape[0])])) for i, image in enumerate(images_64x64): image.save(f"{output_path}/{i}.png") if postprocess_params.get("upsample_scale", False): from swinir import load_model, upsample4x, upsample16x swinir = load_model(postprocess_params.get("swinir_path", "")) upsample_fun = upsample4x if args.upsample_scale == 4 else upsample16x images = upsample_fun(images, swinir).to("cpu") images = list(map(to_pil, [images[i] for i in range(images.shape[0])])) for i, image in enumerate(images): image.save(f"{output_path}/{i}_{args.upsample_scale}x.png") print(f"Images have been saved under {output_path}.") return model_output_dict def parse_args(): import argparse parser = argparse.ArgumentParser() parser.add_argument("--config_file", type=str, default="configs/dalle2_config.py") parser.add_argument("--data_parallel", type=int, default=1) parser.add_argument("--tensor_parallel", type=int, default=4) parser.add_argument("--pipeline_parallel", type=int, default=1) parser.add_argument( "--upsample_scale", type=int, choices=[4, 16], default=None, help="upsample scale, if 4x, output resolution will be 256 x 256.", ) parser.add_argument( "--swinir_path", type=str, default="./swinir/weights/003_realSR_BSRGAN_DFOWMFC_s64w8_SwinIR-L_x4_GAN.pth", ) parser.add_argument("--output_dir", type=str, default="./outputs") parser.add_argument("--save_images", action="store_true") return parser.parse_args() if __name__ == "__main__": args = parse_args() model = Dalle2Pipeline( config_file=args.config_file, data_parallel=args.data_parallel, tensor_parallel=args.tensor_parallel, pipeline_parallel=args.pipeline_parallel, ) texts = [ "a shiba inu wearing a beret and black turtleneck", "a teddy bear on a skateboard in times square", "trump fight with biden in white house", "Donald trump fight with biden in white house", ] imgs = model(texts, **vars(args))
6,770
35.6
99
py
libai
libai-main/projects/DALLE2/dalle2/_clip.py
import os import sys from collections import namedtuple import oneflow as flow from oneflow import nn from .models import l2norm def import_flow_clip(fn): def wrapper(*args, **kwargs): sys.path.append( os.path.join(os.path.abspath(os.path.join(os.path.dirname(__file__), "../..")), "CLIP") ) fn(*args, **kwargs) sys.path.pop() return wrapper EmbeddedText = namedtuple("EmbedTextReturn", ["text_embed", "text_encodings"]) EmbeddedImage = namedtuple("EmbedImageReturn", ["image_embed", "image_encodings"]) class BaseClipAdapter(nn.Module): def __init__(self, clip, **kwargs): super().__init__() self.clip = clip self.overrides = kwargs @property def dim_latent(self): raise NotImplementedError @property def image_size(self): raise NotImplementedError @property def image_channels(self): raise NotImplementedError @property def max_text_len(self): raise NotImplementedError def embed_text(self, text): raise NotImplementedError def embed_image(self, image): raise NotImplementedError class OpenAIClipAdapter(BaseClipAdapter): @import_flow_clip def __init__(self, name="ViT-L/14"): import clip openai_clip, preprocess = clip.load(name) super().__init__(openai_clip) self.eos_id = 49407 # for handling 0 being also '!' text_attention_final = self.find_layer("ln_final") self.handle = text_attention_final.register_forward_hook(self._hook) self.clip_normalize = preprocess.transforms[-1] self.cleared = False def find_layer(self, layer): modules = dict([*self.clip.named_modules()]) return modules.get(layer, None) def clear(self): if self.cleared: return self.handle() def _hook(self, _, inputs, outputs): self.text_encodings = outputs @property def dim_latent(self): return 512 @property def image_size(self): return self.clip.visual.input_resolution @property def image_channels(self): return 3 @property def max_text_len(self): return self.clip.context_length @flow.no_grad() def embed_text(self, text): text = text[..., : self.max_text_len] assert not self.cleared text_mask = text != 0 # v0.15.4 text_embed = self.clip.encode_text(text) text_encodings = self.text_encodings del self.text_encodings return l2norm(text_embed.float()), text_encodings.float(), text_mask @flow.no_grad() def embed_image(self, image): assert not self.cleared image = self.validate_and_resize_image(image) image = self.clip_normalize(image) image_embed = self.clip.encode_image(image) return EmbeddedImage(l2norm(image_embed.float()), None)
2,932
24.068376
99
py
libai
libai-main/projects/DALLE2/dalle2/utils.py
import importlib import time # helper functions def exists(val): return val is not None # time helpers class Timer: def __init__(self): self.reset() def reset(self): self.last_time = time.time() def elapsed(self): return time.time() - self.last_time # print helpers def print_ribbon(s, symbol="=", repeat=40): flank = symbol * repeat return f"{flank} {s} {flank}" # import helpers def import_or_print_error(pkg_name, err_str=None): try: return importlib.import_module(pkg_name) except ModuleNotFoundError: if exists(err_str): print(err_str) exit()
657
14.302326
50
py
libai
libai-main/projects/DALLE2/dalle2/vqgan_vae.py
import copy from functools import partial, wraps from math import sqrt import flowvision import oneflow as flow import oneflow.nn.functional as F from einops import rearrange, repeat from oneflow import einsum, nn from oneflow.autograd import grad as flow_grad from libai.layers import Linear from .einops_exts import Rearrange, rearrange_many from .vector_quantize_flow import VectorQuantize as VQ # constants MList = nn.ModuleList # helper functions def exists(val): return val is not None def default(val, d): return val if exists(val) else d # decorators def eval_decorator(fn): def inner(model, *args, **kwargs): was_training = model.training model.eval() out = fn(model, *args, **kwargs) model.train(was_training) return out return inner def remove_vgg(fn): @wraps(fn) def inner(self, *args, **kwargs): has_vgg = hasattr(self, "vgg") if has_vgg: vgg = self.vgg delattr(self, "vgg") out = fn(self, *args, **kwargs) if has_vgg: self.vgg = vgg return out return inner # keyword argument helpers def pick_and_pop(keys, d): values = list(map(lambda key: d.pop(key), keys)) return dict(zip(keys, values)) def group_dict_by_key(cond, d): return_val = [dict(), dict()] for key in d.keys(): match = bool(cond(key)) ind = int(not match) return_val[ind][key] = d[key] return (*return_val,) def string_begins_with(prefix, string_input): return string_input.startswith(prefix) def group_by_key_prefix(prefix, d): return group_dict_by_key(partial(string_begins_with, prefix), d) def groupby_prefix_and_trim(prefix, d): kwargs_with_prefix, kwargs = group_dict_by_key(partial(string_begins_with, prefix), d) kwargs_without_prefix = dict( map(lambda x: (x[0][len(prefix) :], x[1]), tuple(kwargs_with_prefix.items())) ) return kwargs_without_prefix, kwargs # tensor helper functions def log(t, eps=1e-10): return flow.log(t + eps) def gradient_penalty(images, output, weight=10): images.shape[0] gradients = flow_grad( outputs=output, inputs=images, grad_outputs=flow.ones(output.size(), device=images.device), create_graph=True, retain_graph=True, only_inputs=True, )[0] gradients = rearrange(gradients, "b ... -> b (...)") return weight * ((gradients.norm(2, dim=1) - 1) ** 2).mean() def l2norm(t): return F.normalize(t, dim=-1) def leaky_relu(p=0.1): return nn.LeakyReLU(0.1) def stable_softmax(t, dim=-1, alpha=32 ** 2): t = t / alpha t = t - flow.amax(t, dim=dim, keepdim=True).detach() return (t * alpha).softmax(dim=dim) def safe_div(numer, denom, eps=1e-8): return numer / (denom + eps) # gan losses def hinge_discr_loss(fake, real): return (F.relu(1 + fake) + F.relu(1 - real)).mean() def hinge_gen_loss(fake): return -fake.mean() def bce_discr_loss(fake, real): return (-log(1 - flow.sigmoid(fake)) - log(flow.sigmoid(real))).mean() def bce_gen_loss(fake): return -log(flow.sigmoid(fake)).mean() def grad_layer_wrt_loss(loss, layer): return flow_grad( outputs=loss, inputs=layer, grad_outputs=flow.ones_like(loss), retain_graph=True )[0].detach() # vqgan vae class LayerNormChan(nn.Module): def __init__(self, dim, eps=1e-5): super().__init__() self.eps = eps self.gamma = nn.Parameter(flow.ones(1, dim, 1, 1)) def forward(self, x): var = flow.var(x, dim=1, unbiased=False, keepdim=True) mean = flow.mean(x, dim=1, keepdim=True) return (x - mean) / (var + self.eps).sqrt() * self.gamma # discriminator class Discriminator(nn.Module): def __init__(self, dims, channels=3, groups=16, init_kernel_size=5): super().__init__() dim_pairs = zip(dims[:-1], dims[1:]) self.layers = MList( [ nn.Sequential( nn.Conv2d(channels, dims[0], init_kernel_size, padding=init_kernel_size // 2), leaky_relu(), ) ] ) for dim_in, dim_out in dim_pairs: self.layers.append( nn.Sequential( nn.Conv2d(dim_in, dim_out, 4, stride=2, padding=1), nn.GroupNorm(groups, dim_out), leaky_relu(), ) ) dim = dims[-1] self.to_logits = nn.Sequential( # return 5 x 5, for PatchGAN-esque training nn.Conv2d(dim, dim, 1), leaky_relu(), nn.Conv2d(dim, 1, 4) ) def forward(self, x): for net in self.layers: x = net(x) return self.to_logits(x) # positional encoding class ContinuousPositionBias(nn.Module): """from https://arxiv.org/abs/2111.09883""" def __init__(self, *, dim, heads, layers=2): super().__init__() self.net = MList([]) self.net.append(nn.Sequential(Linear(2, dim), leaky_relu())) for _ in range(layers - 1): self.net.append(nn.Sequential(Linear(dim, dim), leaky_relu())) self.net.append(Linear(dim, heads)) self.register_buffer("rel_pos", None, persistent=False) def forward(self, x): n, device = x.shape[-1], x.device fmap_size = int(sqrt(n)) if not exists(self.rel_pos): pos = flow.arange(fmap_size, device=device) grid = flow.stack(flow.meshgrid(pos, pos, indexing="ij")) grid = rearrange(grid, "c i j -> (i j) c") rel_pos = rearrange(grid, "i c -> i 1 c") - rearrange(grid, "j c -> 1 j c") rel_pos = flow.sign(rel_pos) * flow.log(rel_pos.abs() + 1) self.register_buffer("rel_pos", rel_pos, persistent=False) rel_pos = self.rel_pos.float() for layer in self.net: rel_pos = layer(rel_pos) bias = rearrange(rel_pos, "i j h -> h i j") return x + bias # resnet encoder / decoder class ResnetEncDec(nn.Module): def __init__( self, dim, *, channels=3, layers=4, layer_mults=None, num_resnet_blocks=1, resnet_groups=16, first_conv_kernel_size=5, use_attn=True, attn_dim_head=64, attn_heads=8, attn_dropout=0.0, ): super().__init__() assert ( dim % resnet_groups == 0 ), f"dimension {dim} must be divisible by {resnet_groups} (groups for the groupnorm)" self.layers = layers self.encoders = MList([]) self.decoders = MList([]) layer_mults = default(layer_mults, list(map(lambda t: 2 ** t, range(layers)))) assert ( len(layer_mults) == layers ), "layer multipliers must be equal to designated number of layers" layer_dims = [dim * mult for mult in layer_mults] dims = (dim, *layer_dims) self.encoded_dim = dims[-1] dim_pairs = zip(dims[:-1], dims[1:]) def append(arr, t): arr.append(t) def prepend(arr, t): arr.insert(0, t) if not isinstance(num_resnet_blocks, tuple): num_resnet_blocks = (*((0,) * (layers - 1)), num_resnet_blocks) if not isinstance(use_attn, tuple): use_attn = (*((False,) * (layers - 1)), use_attn) assert ( len(num_resnet_blocks) == layers ), "number of resnet blocks config must be equal to number of layers" assert len(use_attn) == layers for layer_index, (dim_in, dim_out), layer_num_resnet_blocks, layer_use_attn in zip( range(layers), dim_pairs, num_resnet_blocks, use_attn ): append( self.encoders, nn.Sequential(nn.Conv2d(dim_in, dim_out, 4, stride=2, padding=1), leaky_relu()), ) prepend( self.decoders, nn.Sequential(nn.ConvTranspose2d(dim_out, dim_in, 4, 2, 1), leaky_relu()), ) if layer_use_attn: prepend( self.decoders, VQGanAttention( dim=dim_out, heads=attn_heads, dim_head=attn_dim_head, dropout=attn_dropout ), ) for _ in range(layer_num_resnet_blocks): append(self.encoders, ResBlock(dim_out, groups=resnet_groups)) prepend(self.decoders, GLUResBlock(dim_out, groups=resnet_groups)) if layer_use_attn: append( self.encoders, VQGanAttention( dim=dim_out, heads=attn_heads, dim_head=attn_dim_head, dropout=attn_dropout ), ) prepend( self.encoders, nn.Conv2d(channels, dim, first_conv_kernel_size, padding=first_conv_kernel_size // 2), ) append(self.decoders, nn.Conv2d(dim, channels, 1)) def get_encoded_fmap_size(self, image_size): return image_size // (2 ** self.layers) @property def last_dec_layer(self): return self.decoders[-1].weight def encode(self, x): for enc in self.encoders: x = enc(x) return x def decode(self, x): for dec in self.decoders: x = dec(x) return x class GLUResBlock(nn.Module): def __init__(self, chan, groups=16): super().__init__() self.net = nn.Sequential( nn.Conv2d(chan, chan * 2, 3, padding=1), nn.GLU(dim=1), nn.GroupNorm(groups, chan), nn.Conv2d(chan, chan * 2, 3, padding=1), nn.GLU(dim=1), nn.GroupNorm(groups, chan), nn.Conv2d(chan, chan, 1), ) def forward(self, x): return self.net(x) + x class ResBlock(nn.Module): def __init__(self, chan, groups=16): super().__init__() self.net = nn.Sequential( nn.Conv2d(chan, chan, 3, padding=1), nn.GroupNorm(groups, chan), leaky_relu(), nn.Conv2d(chan, chan, 3, padding=1), nn.GroupNorm(groups, chan), leaky_relu(), nn.Conv2d(chan, chan, 1), ) def forward(self, x): return self.net(x) + x # vqgan attention layer class VQGanAttention(nn.Module): def __init__(self, *, dim, dim_head=64, heads=8, dropout=0.0): super().__init__() self.heads = heads self.scale = dim_head ** -0.5 inner_dim = heads * dim_head self.dropout = nn.Dropout(dropout) self.pre_norm = LayerNormChan(dim) self.cpb = ContinuousPositionBias(dim=dim // 4, heads=heads) self.to_qkv = nn.Conv2d(dim, inner_dim * 3, 1, bias=False) self.to_out = nn.Conv2d(inner_dim, dim, 1, bias=False) def forward(self, x): h = self.heads height, width, residual = *x.shape[-2:], x.clone() x = self.pre_norm(x) q, k, v = self.to_qkv(x).chunk(3, dim=1) q, k, v = map(lambda t: rearrange(t, "b (h c) x y -> b h c (x y)", h=h), (q, k, v)) sim = einsum("b h c i, b h c j -> b h i j", q, k) * self.scale sim = self.cpb(sim) attn = stable_softmax(sim, dim=-1) attn = self.dropout(attn) out = einsum("b h i j, b h c j -> b h c i", attn, v) out = rearrange(out, "b h c (x y) -> b (h c) x y", x=height, y=width) out = self.to_out(out) return out + residual # ViT encoder / decoder class RearrangeImage(nn.Module): def forward(self, x): n = x.shape[1] w = h = int(sqrt(n)) return rearrange(x, "b (h w) ... -> b h w ...", h=h, w=w) class Attention(nn.Module): def __init__(self, dim, *, heads=8, dim_head=32): super().__init__() self.norm = nn.LayerNorm(dim) self.heads = heads self.scale = dim_head ** -0.5 inner_dim = dim_head * heads self.to_qkv = nn.Linear(dim, inner_dim * 3, bias=False) self.to_out = nn.Linear(inner_dim, dim) def forward(self, x): h = self.heads x = self.norm(x) q, k, v = self.to_qkv(x).chunk(3, dim=-1) q, k, v = rearrange_many((q, k, v), "b n (h d) -> b h n d", h=h) q = q * self.scale sim = einsum("b h i d, b h j d -> b h i j", q, k) sim = sim - sim.amax(dim=-1, keepdim=True).detach() attn = sim.softmax(dim=-1) out = einsum("b h i j, b h j d -> b h i d", attn, v) out = rearrange(out, "b h n d -> b n (h d)") return self.to_out(out) def FeedForward(dim, mult=4): return nn.Sequential( nn.LayerNorm(dim), nn.Linear(dim, dim * mult, bias=False), nn.GELU(), nn.Linear(dim * mult, dim, bias=False), ) class Transformer(nn.Module): def __init__(self, dim, *, layers, dim_head=32, heads=8, ff_mult=4): super().__init__() self.layers = nn.ModuleList([]) for _ in range(layers): self.layers.append( nn.ModuleList( [ Attention(dim=dim, dim_head=dim_head, heads=heads), FeedForward(dim=dim, mult=ff_mult), ] ) ) self.norm = nn.LayerNorm(dim) def forward(self, x): for attn, ff in self.layers: x = attn(x) + x x = ff(x) + x return self.norm(x) class ViTEncDec(nn.Module): def __init__(self, dim, channels=3, layers=4, patch_size=8, dim_head=32, heads=8, ff_mult=4): super().__init__() self.encoded_dim = dim self.patch_size = patch_size input_dim = channels * (patch_size ** 2) self.encoder = nn.Sequential( Rearrange("b c (h p1) (w p2) -> b (h w) (p1 p2 c)", p1=patch_size, p2=patch_size), Linear(input_dim, dim), Transformer(dim=dim, dim_head=dim_head, heads=heads, ff_mult=ff_mult, layers=layers), RearrangeImage(), Rearrange("b h w c -> b c h w"), ) self.decoder = nn.Sequential( Rearrange("b c h w -> b (h w) c"), Transformer(dim=dim, dim_head=dim_head, heads=heads, ff_mult=ff_mult, layers=layers), nn.Sequential( Linear(dim, dim * 4, bias=False), nn.Tanh(), Linear(dim * 4, input_dim, bias=False), ), RearrangeImage(), Rearrange("b h w (p1 p2 c) -> b c (h p1) (w p2)", p1=patch_size, p2=patch_size), ) def get_encoded_fmap_size(self, image_size): return image_size // self.patch_size @property def last_dec_layer(self): return self.decoder[-3][-1].weight def encode(self, x): return self.encoder(x) def decode(self, x): return self.decoder(x) # main vqgan-vae classes class NullVQGanVAE(nn.Module): def __init__(self, *, channels): super().__init__() self.encoded_dim = channels self.layers = 0 def get_encoded_fmap_size(self, size): return size def copy_for_eval(self): return self def encode(self, x): return x def decode(self, x): return x class VQGanVAE(nn.Module): def __init__( self, *, dim, image_size, channels=3, layers=4, l2_recon_loss=False, use_hinge_loss=True, vgg=None, vq_codebook_dim=256, vq_codebook_size=512, vq_decay=0.8, vq_commitment_weight=1.0, vq_kmeans_init=True, vq_use_cosine_sim=True, use_vgg_and_gan=True, vae_type="resnet", discr_layers=4, **kwargs, ): super().__init__() vq_kwargs, kwargs = groupby_prefix_and_trim("vq_", kwargs) encdec_kwargs, kwargs = groupby_prefix_and_trim("encdec_", kwargs) self.image_size = image_size self.channels = channels self.codebook_size = vq_codebook_size if vae_type == "resnet": enc_dec_klass = ResnetEncDec elif vae_type == "vit": enc_dec_klass = ViTEncDec else: raise ValueError(f"{vae_type} not valid") self.enc_dec = enc_dec_klass(dim=dim, channels=channels, layers=layers, **encdec_kwargs) self.vq = VQ( dim=self.enc_dec.encoded_dim, codebook_dim=vq_codebook_dim, codebook_size=vq_codebook_size, decay=vq_decay, commitment_weight=vq_commitment_weight, accept_image_fmap=True, kmeans_init=vq_kmeans_init, use_cosine_sim=vq_use_cosine_sim, **vq_kwargs, ) # reconstruction loss self.recon_loss_fn = F.mse_loss if l2_recon_loss else F.l1_loss # turn off GAN and perceptual loss if grayscale self.vgg = None self.discr = None self.use_vgg_and_gan = use_vgg_and_gan if not use_vgg_and_gan: return # preceptual loss if exists(vgg): self.vgg = vgg else: self.vgg = flowvision.models.vgg16(pretrained=True) self.vgg.classifier = nn.Sequential(*self.vgg.classifier[:-2]) # gan related losses layer_mults = list(map(lambda t: 2 ** t, range(discr_layers))) layer_dims = [dim * mult for mult in layer_mults] dims = (dim, *layer_dims) self.discr = Discriminator(dims=dims, channels=channels) self.discr_loss = hinge_discr_loss if use_hinge_loss else bce_discr_loss self.gen_loss = hinge_gen_loss if use_hinge_loss else bce_gen_loss @property def encoded_dim(self): return self.enc_dec.encoded_dim def get_encoded_fmap_size(self, image_size): return self.enc_dec.get_encoded_fmap_size(image_size) def copy_for_eval(self): device = next(self.parameters()).device vae_copy = copy.deepcopy(self.cpu()) if vae_copy.use_vgg_and_gan: del vae_copy.discr del vae_copy.vgg vae_copy.eval() return vae_copy.to(device) @remove_vgg def state_dict(self, *args, **kwargs): return super().state_dict(*args, **kwargs) @remove_vgg def load_state_dict(self, *args, **kwargs): return super().load_state_dict(*args, **kwargs) @property def codebook(self): return self.vq.codebook def encode(self, fmap): fmap = self.enc_dec.encode(fmap) return fmap def decode(self, fmap, return_indices_and_loss=False): fmap, indices, commit_loss = self.vq(fmap) fmap = self.enc_dec.decode(fmap) if not return_indices_and_loss: return fmap return fmap, indices, commit_loss def forward( self, img, return_loss=False, return_discr_loss=False, return_recons=False, add_gradient_penalty=True, ): _, channels, height, width, _ = *img.shape, img.device assert ( height == self.image_size and width == self.image_size ), "height and width of input image must be equal to {self.image_size}" assert ( channels == self.channels ), "number of channels on image or sketch is not equal to the channels set on this VQGanVAE" fmap = self.encode(img) fmap, indices, commit_loss = self.decode(fmap, return_indices_and_loss=True) if not return_loss and not return_discr_loss: return fmap assert ( return_loss ^ return_discr_loss ), "you should either return autoencoder loss or discriminator loss, but not both" # whether to return discriminator loss if return_discr_loss: assert exists(self.discr), "discriminator must exist to train it" fmap.detach_() img.requires_grad_() fmap_discr_logits, img_discr_logits = map(self.discr, (fmap, img)) discr_loss = self.discr_loss(fmap_discr_logits, img_discr_logits) if add_gradient_penalty: gp = gradient_penalty(img, img_discr_logits) loss = discr_loss + gp if return_recons: return loss, fmap return loss # reconstruction loss recon_loss = self.recon_loss_fn(fmap, img) # early return if training on grayscale if not self.use_vgg_and_gan: if return_recons: return recon_loss, fmap return recon_loss # perceptual loss img_vgg_input = img fmap_vgg_input = fmap if img.shape[1] == 1: # handle grayscale for vgg img_vgg_input, fmap_vgg_input = map( lambda t: repeat(t, "b 1 ... -> b c ...", c=3), (img_vgg_input, fmap_vgg_input) ) img_vgg_feats = self.vgg(img_vgg_input) recon_vgg_feats = self.vgg(fmap_vgg_input) perceptual_loss = F.mse_loss(img_vgg_feats, recon_vgg_feats) # generator loss gen_loss = self.gen_loss(self.discr(fmap)) # calculate adaptive weight last_dec_layer = self.enc_dec.last_dec_layer norm_grad_wrt_gen_loss = grad_layer_wrt_loss(gen_loss, last_dec_layer).norm(p=2) norm_grad_wrt_perceptual_loss = grad_layer_wrt_loss(perceptual_loss, last_dec_layer).norm( p=2 ) adaptive_weight = safe_div(norm_grad_wrt_perceptual_loss, norm_grad_wrt_gen_loss) adaptive_weight.clamp_(max=1e4) # combine losses loss = recon_loss + perceptual_loss + commit_loss + adaptive_weight * gen_loss if return_recons: return loss, fmap return loss
21,938
26.219603
100
py
libai
libai-main/projects/DALLE2/dalle2/rotary_embedding_flow.py
from inspect import isfunction from math import pi import oneflow as flow from einops import rearrange, repeat from oneflow import einsum, nn from libai.utils import distributed as dist # helper functions def exists(val): return val is not None def broadcat(tensors, dim=-1): num_tensors = len(tensors) shape_lens = set(list(map(lambda t: len(t.shape), tensors))) assert len(shape_lens) == 1, "tensors must all have the same number of dimensions" shape_len = list(shape_lens)[0] dim = (dim + shape_len) if dim < 0 else dim dims = list(zip(*map(lambda t: list(t.shape), tensors))) expandable_dims = [(i, val) for i, val in enumerate(dims) if i != dim] assert all( [*map(lambda t: len(set(t[1])) <= 2, expandable_dims)] ), "invalid dimensions for broadcastable concatentation" max_dims = list(map(lambda t: (t[0], max(t[1])), expandable_dims)) expanded_dims = list(map(lambda t: (t[0], (t[1],) * num_tensors), max_dims)) expanded_dims.insert(dim, (dim, dims[dim])) expandable_shapes = list(zip(*map(lambda t: t[1], expanded_dims))) tensors = list(map(lambda t: t[0].expand(*t[1]), zip(tensors, expandable_shapes))) return flow.cat(tensors, dim=dim) # rotary embedding helper functions def rotate_half(x): x = rearrange(x, "... (d r) -> ... d r", r=2) x1, x2 = [i.squeeze(-1) for i in x.chunk(2, -1)] # x.unbind(dim = -1) x = flow.stack((-x2, x1), dim=-1) return rearrange(x, "... d r -> ... (d r)") def apply_rotary_emb(freqs, t, start_index=0): freqs = freqs.to(t.dtype) rot_dim = freqs.shape[-1] end_index = start_index + rot_dim assert ( rot_dim <= t.shape[-1] ), f"feature dimension {t.shape[-1]} is not of sufficient size to \ rotate in all the positions {rot_dim}" t_left, t, t_right = t[..., :start_index], t[..., start_index:end_index], t[..., end_index:] t = (t * freqs.cos()) + (rotate_half(t) * freqs.sin()) return flow.cat((t_left, t, t_right), dim=-1) # learned rotation helpers def apply_learned_rotations(rotations, t, start_index=0, freq_ranges=None): if exists(freq_ranges): rotations = einsum("..., f -> ... f", rotations, freq_ranges) rotations = rearrange(rotations, "... r f -> ... (r f)") rotations = repeat(rotations, "... n -> ... (n r)", r=2) return apply_rotary_emb(rotations, t, start_index=start_index) # classes class RotaryEmbedding(nn.Module): def __init__( self, dim, custom_freqs=None, freqs_for="lang", theta=10000, max_freq=10, num_freqs=1, learned_freq=False, ): super().__init__() if exists(custom_freqs): freqs = custom_freqs elif freqs_for == "lang": freqs = 1.0 / (theta ** (flow.arange(0, dim, 2)[: (dim // 2)].float() / dim)) elif freqs_for == "pixel": freqs = flow.linspace(1.0, max_freq / 2, dim // 2) * pi elif freqs_for == "constant": freqs = flow.ones(num_freqs).float() else: raise ValueError(f"unknown modality {freqs_for}") self.cache = dict() if learned_freq: self.freqs = nn.Parameter(freqs) else: self.register_buffer("freqs", freqs) def rotate_queries_or_keys(self, t, seq_dim=-2): seq_len = t.shape[seq_dim] freqs = self.forward( lambda: flow.arange( seq_len, placement=dist.get_layer_placement(0), sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), ), cache_key=seq_len, ) return apply_rotary_emb(freqs, t) def forward(self, t, cache_key=None): if exists(cache_key) and cache_key in self.cache: return self.cache[cache_key] if isfunction(t): t = t() freqs = self.freqs freqs = flow.einsum("..., f -> ... f", t.to(freqs.dtype), freqs) freqs = repeat(freqs, "... n -> ... (n r)", r=2) if exists(cache_key): self.cache[cache_key] = freqs return freqs
4,157
30.029851
96
py
libai
libai-main/projects/DALLE2/dalle2/tokenizer.py
# take from https://github.com/openai/CLIP/blob/main/clip/simple_tokenizer.py # to give users a quick easy start to training DALL-E without doing BPE import gzip import html import os from functools import lru_cache from pathlib import Path import ftfy import oneflow as flow import regex as re from .utils import import_or_print_error # OpenAI simple tokenizer @lru_cache() def default_bpe(): return os.path.join( os.path.dirname(os.path.abspath(__file__)), "data/bpe_simple_vocab_16e6.txt.gz" ) @lru_cache() def bytes_to_unicode(): bs = ( list(range(ord("!"), ord("~") + 1)) + list(range(ord("¡"), ord("¬") + 1)) + list(range(ord("®"), ord("ÿ") + 1)) ) cs = bs[:] n = 0 for b in range(2 ** 8): if b not in bs: bs.append(b) cs.append(2 ** 8 + n) n += 1 cs = [chr(n) for n in cs] return dict(zip(bs, cs)) def get_pairs(word): pairs = set() prev_char = word[0] for char in word[1:]: pairs.add((prev_char, char)) prev_char = char return pairs def basic_clean(text): text = ftfy.fix_text(text) text = html.unescape(html.unescape(text)) return text.strip() def whitespace_clean(text): text = re.sub(r"\s+", " ", text) text = text.strip() return text class SimpleTokenizer(object): def __init__(self, bpe_path=default_bpe()): self.byte_encoder = bytes_to_unicode() self.byte_decoder = {v: k for k, v in self.byte_encoder.items()} merges = ( gzip.open(bpe_path).read().decode("utf-8").split("\n") ) # Path(bpe_path).read_text(encoding='utf8').split('\n') merges = merges[1 : 49152 - 256 - 2 + 1] merges = [tuple(merge.split()) for merge in merges] vocab = list(bytes_to_unicode().values()) vocab = vocab + [v + "</w>" for v in vocab] for merge in merges: vocab.append("".join(merge)) vocab.extend(["<|startoftext|>", "<|endoftext|>"]) self.vocab_size = 49408 self.encoder = dict(zip(vocab, range(len(vocab)))) self.decoder = {v: k for k, v in self.encoder.items()} self.bpe_ranks = dict(zip(merges, range(len(merges)))) self.cache = {"<|startoftext|>": "<|startoftext|>", "<|endoftext|>": "<|endoftext|>"} self.pat = re.compile( r"""<\|startoftext\|>|<\|endoftext\|>|'s|'t|'re|'ve|'m|'ll|'d|[\p{L}]+|[\p{N}]|[^\s\p{L}\p{N}]+""", # noqa re.IGNORECASE, ) def bpe(self, token): if token in self.cache: return self.cache[token] word = tuple(token[:-1]) + (token[-1] + "</w>",) pairs = get_pairs(word) if not pairs: return token + "</w>" while True: bigram = min(pairs, key=lambda pair: self.bpe_ranks.get(pair, float("inf"))) if bigram not in self.bpe_ranks: break first, second = bigram new_word = [] i = 0 while i < len(word): try: j = word.index(first, i) new_word.extend(word[i:j]) i = j except ValueError: new_word.extend(word[i:]) break if word[i] == first and i < len(word) - 1 and word[i + 1] == second: new_word.append(first + second) i += 2 else: new_word.append(word[i]) i += 1 new_word = tuple(new_word) word = new_word if len(word) == 1: break else: pairs = get_pairs(word) word = " ".join(word) self.cache[token] = word return word def encode(self, text): bpe_tokens = [] text = whitespace_clean(basic_clean(text)).lower() for token in re.findall(self.pat, text): token = "".join(self.byte_encoder[b] for b in token.encode("utf-8")) bpe_tokens.extend(self.encoder[bpe_token] for bpe_token in self.bpe(token).split(" ")) return bpe_tokens def decode(self, tokens, remove_start_end=True, pad_tokens=set()): if flow.is_tensor(tokens): tokens = tokens.tolist() if remove_start_end: tokens = [token for token in tokens if token not in (49406, 40407, 0)] text = "".join([self.decoder[token] for token in tokens if token not in pad_tokens]) text = ( bytearray([self.byte_decoder[c] for c in text]) .decode("utf-8", errors="replace") .replace("</w>", " ") ) return text def tokenize(self, texts, context_length=256, truncate_text=False): if isinstance(texts, str): texts = [texts] all_tokens = [self.encode(text) for text in texts] result = flow.zeros(len(all_tokens), context_length, dtype=flow.long) for i, tokens in enumerate(all_tokens): if len(tokens) > context_length: if truncate_text: tokens = tokens[:context_length] else: raise RuntimeError( f"Input {texts[i]} is too long for context length {context_length}" ) result[i, : len(tokens)] = flow.tensor(tokens) return result # tokenizer = SimpleTokenizer() # YTTM tokenizer class YttmTokenizer: def __init__(self, bpe_path=None): bpe_path = Path(bpe_path) assert bpe_path.exists(), f"BPE json path {str(bpe_path)} does not exist" self.yttm = import_or_print_error( "youtokentome", "you need to install youtokentome by `pip install youtokentome`" ) tokenizer = self.yttm.BPE(model=str(bpe_path)) self.tokenizer = tokenizer self.vocab_size = tokenizer.vocab_size() def decode(self, tokens, pad_tokens=set()): if flow.is_tensor(tokens): tokens = tokens.tolist() return self.tokenizer.decode(tokens, ignore_ids=pad_tokens.union({0})) def encode(self, texts): encoded = self.tokenizer.encode(texts, output_type=self.yttm.OutputType.ID) return list(map(flow.tensor, encoded)) def tokenize(self, texts, context_length=256, truncate_text=False): if isinstance(texts, str): texts = [texts] all_tokens = self.encode(texts) result = flow.zeros(len(all_tokens), context_length, dtype=flow.long) for i, tokens in enumerate(all_tokens): if len(tokens) > context_length: if truncate_text: tokens = tokens[:context_length] else: raise RuntimeError( f"Input {texts[i]} is too long for context length {context_length}" ) result[i, : len(tokens)] = flow.tensor(tokens) return result
7,011
30.872727
119
py
libai
libai-main/projects/DALLE2/dalle2/dalle2_loader.py
import logging import oneflow as flow from oneflow.framework.check_point_v2 import _broadcast_py_object import libai.utils.distributed as dist from libai.models.build import build_model from libai.models.utils.model_loader.base_loader import ( ModelLoaderHuggerFace, _load_state_dict_into_model, ) logger = logging.getLogger("libai.dalle2." + __name__) class Dalle2ModelLoader(ModelLoaderHuggerFace): def __init__(self, model, libai_cfg, pretrained_model_path, **kwargs): super().__init__(model, libai_cfg, pretrained_model_path, **kwargs) self.base_model_prefix_1 = "" self.base_model_prefix_2 = "" def _convert_state_dict(self, state_dict, module="prior"): old_keys = [] new_keys = [] if module == "prior": for k in state_dict.keys(): if "clip." in k: continue old_keys.append(k) if k.endswith(".g"): k = k[:-1] + "weight" elif k.startswith("net.causal_transformer"): if k.endswith("gamma"): k = k[:-5] + "weight" elif k.endswith("beta"): k = k[:-4] + "bias" new_keys.append("prior." + k) elif module == "decoder": for k in state_dict.keys(): if "clip." in k: continue old_keys.append(k) if k.endswith(".g"): k = k[:-1] + "weight" elif "cross_attn" in k: if k.endswith("gamma"): k = k[:-5] + "weight" elif k.endswith("beta"): k = k[:-4] + "bias" new_keys.append("decoder." + k) ret_state_dict = {} for old_key, new_key in zip(old_keys, new_keys): ret_state_dict[new_key] = state_dict.pop(old_key) return ret_state_dict def load(self): if dist.is_main_process(): # prior logger.info("loading torch model prior...") torch_state_dict = self._load_torch_state_dict(self.libai_cfg.model.prior_weight_path)[ "ema_model" ] logger.info("converting torch model prior into oneflow model...") flow_state_dict = self._convert_tensors(torch_state_dict) prior_state_dict = self._convert_state_dict(flow_state_dict) # decoder logger.info("loading torch model decoder...") torch_state_dict = self._load_torch_state_dict(self.libai_cfg.model.decoder_weight_path) flow_state_dict = self._convert_tensors(torch_state_dict) logger.info("converting torch model decoder into oneflow model...") decoder_state_dict = self._convert_state_dict(flow_state_dict, module="decoder") flow_state_dict = {**prior_state_dict, **decoder_state_dict} else: flow_state_dict = None logger.info("building LiBai model...") self.libai_cfg = _broadcast_py_object(self.libai_cfg, src=0) self.model = build_model(self.model) self.model._apply(dist.convert_to_distributed_default_setting) self.model = self.model.eval() flow.cuda.empty_cache() # State_dict to global logger.info("transfering state_dict local to global...") flow_state_dict = self._state_dict_to_global(flow_state_dict, mode="pytorch") # oom # Load # ( # model, # missing_keys, # unexpected_keys, # mismatched_keys, # error_msgs, # ) = self._load_pretrained_model(self.model, flow_state_dict, self.pretrained_model_path) logger.info("loading model weights into LiBai...") _load_state_dict_into_model(self.model, flow_state_dict, "") return self.model
3,927
39.494845
100
py
libai
libai-main/projects/DALLE2/dalle2/models.py
import math import random from contextlib import contextmanager from functools import partial, wraps import flowvision.transforms as T import kornia.augmentation as K import numpy as np import oneflow as flow import oneflow.nn.functional as F from einops import rearrange, reduce, repeat from kornia.filters import gaussian_blur2d from omegaconf import ListConfig from oneflow import einsum, nn from oneflow.nn import Conv2d, ConvTranspose2d, GroupNorm from oneflow.nn.functional import layer_norm from resize_right import resize from tqdm.auto import tqdm from libai.layers import Embedding, LayerNorm, Linear from libai.utils import distributed as dist from .einops_exts import EinopsToAndFrom, Rearrange, check_shape, rearrange_many, repeat_many from .rotary_embedding_flow import RotaryEmbedding from .tokenizer import SimpleTokenizer from .vqgan_vae import NullVQGanVAE, VQGanVAE # rotary embeddings # constants def get_default_sbp(): return dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]) def get_default_placement(): return dist.get_layer_placement(0) NAT = 1.0 / math.log(2.0) random.seed(666) np.random.seed(6666) # helper functions def exists(val): return val is not None def identity(t, *args, **kwargs): return t def first(arr, d=None): if len(arr) == 0: return d return arr[0] def maybe(fn): @wraps(fn) def inner(x): if not exists(x): return x return fn(x) return inner def default(val, d): if exists(val): return val return d() if callable(d) else d def cast_tuple(val, length=None): if isinstance(val, list) or isinstance(val, ListConfig): val = tuple(val) out = val if isinstance(val, tuple) else ((val,) * default(length, 1)) if exists(length): assert len(out) == length return out def module_device(module): return next(module.parameters()).device def zero_init_(m): nn.init.zeros_(m.weight) if exists(m.bias): nn.init.zeros_(m.bias) @contextmanager def null_context(*args, **kwargs): yield def eval_decorator(fn): def inner(model, *args, **kwargs): was_training = model.training model.eval() out = fn(model, *args, **kwargs) model.train(was_training) return out return inner def is_list_str(x): if not isinstance(x, (list, tuple)): return False return all([type(el) == str for el in x]) def pad_tuple_to_length(t, length, fillvalue=None): remain_length = length - len(t) if remain_length <= 0: return t return (*t, *((fillvalue,) * remain_length)) # for controlling freezing of CLIP def set_module_requires_grad_(module, requires_grad): for param in module.parameters(): param.requires_grad = requires_grad def freeze_all_layers_(module): set_module_requires_grad_(module, False) def unfreeze_all_layers_(module): set_module_requires_grad_(module, True) def freeze_model_and_make_eval_(model): model.eval() freeze_all_layers_(model) # tensor helpers def log(t, eps=1e-12): return flow.log(t.clamp(min=eps)) def l2norm(t): return F.normalize(t, dim=-1) def resize_image_to(image, target_image_size): orig_image_size = image.shape[-1] if orig_image_size == target_image_size: return image scale_factors = target_image_size / orig_image_size return resize(image, scale_factors=scale_factors) # image normalization functions # ddpms expect images to be in the range of -1 to 1 # but CLIP may otherwise def normalize_neg_one_to_one(img): return img * 2 - 1 def unnormalize_zero_to_one(normed_img): return (normed_img + 1) * 0.5 # classifier free guidance functions def prob_mask_like(shape, prob, placement=None, sbp=None): placement = placement or get_default_placement() sbp = sbp or get_default_sbp() if prob == 1: return flow.ones(shape, dtype=flow.bool, placement=placement, sbp=sbp) elif prob == 0: return flow.zeros(shape, dtype=flow.bool, placement=placement, sbp=sbp) else: return flow.zeros(shape, placement=placement, sbp=sbp).float().uniform_(0, 1) < prob # gaussian diffusion helper functions def extract(a, t, x_shape): b, *_ = t.shape out = a.gather(len(a.shape) - 1, t) return out.reshape(b, *((1,) * (len(x_shape) - 1))) def meanflat(x): return x.mean(dim=tuple(range(1, len(x.shape)))) def normal_kl(mean1, logvar1, mean2, logvar2): return 0.5 * ( -1.0 + logvar2 - logvar1 + flow.exp(logvar1 - logvar2) + ((mean1 - mean2) ** 2) * flow.exp(-logvar2) ) def approx_standard_normal_cdf(x): return 0.5 * (1.0 + flow.tanh(((2.0 / math.pi) ** 0.5) * (x + 0.044715 * (x ** 3)))) def discretized_gaussian_log_likelihood(x, *, means, log_scales, thres=0.999): assert x.shape == means.shape == log_scales.shape # attempting to correct nan gradients when learned variance is turned on # in the setting of deepspeed fp16 eps = 1e-12 if x.dtype == flow.float32 else 1e-3 centered_x = x - means inv_stdv = flow.exp(-log_scales) plus_in = inv_stdv * (centered_x + 1.0 / 255.0) cdf_plus = approx_standard_normal_cdf(plus_in) min_in = inv_stdv * (centered_x - 1.0 / 255.0) cdf_min = approx_standard_normal_cdf(min_in) log_cdf_plus = log(cdf_plus) log_one_minus_cdf_min = log(1.0 - cdf_min) cdf_delta = cdf_plus - cdf_min log_probs = flow.where( x < -thres, log_cdf_plus, flow.where(x > thres, log_one_minus_cdf_min, log(cdf_delta, eps)) ) return log_probs def cosine_beta_schedule(timesteps, s=0.008): """ cosine schedule as proposed in https://openreview.net/forum?id=-NEXDKk8gZ """ steps = timesteps + 1 x = flow.linspace(0, timesteps, steps, dtype=flow.float64) alphas_cumprod = flow.cos(((x / timesteps) + s) / (1 + s) * np.pi * 0.5) ** 2 alphas_cumprod = alphas_cumprod / first(alphas_cumprod) betas = 1 - (alphas_cumprod[1:] / alphas_cumprod[:-1]) return flow.clip(betas, 0, 0.999) def linear_beta_schedule(timesteps): scale = 1000 / timesteps beta_start = scale * 0.0001 beta_end = scale * 0.02 return flow.linspace(beta_start, beta_end, timesteps, dtype=flow.float64) def quadratic_beta_schedule(timesteps): scale = 1000 / timesteps beta_start = scale * 0.0001 beta_end = scale * 0.02 return flow.linspace(beta_start ** 0.5, beta_end ** 0.5, timesteps, dtype=flow.float64) ** 2 def sigmoid_beta_schedule(timesteps): scale = 1000 / timesteps beta_start = scale * 0.0001 beta_end = scale * 0.02 betas = flow.linspace(-6, 6, timesteps, dtype=flow.float64) return flow.sigmoid(betas) * (beta_end - beta_start) + beta_start class NoiseScheduler(nn.Module): def __init__( self, *, beta_schedule, timesteps, loss_type, p2_loss_weight_gamma=0.0, p2_loss_weight_k=1 ): super().__init__() if beta_schedule == "cosine": betas = cosine_beta_schedule(timesteps) elif beta_schedule == "linear": betas = linear_beta_schedule(timesteps) elif beta_schedule == "quadratic": betas = quadratic_beta_schedule(timesteps) elif beta_schedule == "jsd": betas = 1.0 / flow.linspace(timesteps, 1, timesteps) elif beta_schedule == "sigmoid": betas = sigmoid_beta_schedule(timesteps) else: raise NotImplementedError() betas = betas alphas = 1.0 - betas alphas_cumprod = flow.cumprod(alphas, dim=0) alphas_cumprod_prev = F.pad(alphas_cumprod[:-1], (1, 0), value=1.0) (timesteps,) = betas.shape self.num_timesteps = int(timesteps) if loss_type == "l1": loss_fn = F.l1_loss elif loss_type == "l2": loss_fn = flow.nn.MSELoss() elif loss_type == "huber": loss_fn = F.smooth_l1_loss else: raise NotImplementedError() self.loss_type = loss_type self.loss_fn = loss_fn # register buffer helper function to cast double back to float def register_buffer(name, val): self.register_buffer(name, val.to(flow.float32)) register_buffer("betas", betas) register_buffer("alphas_cumprod", alphas_cumprod) register_buffer("alphas_cumprod_prev", alphas_cumprod_prev) # calculations for diffusion q(x_t | x_{t-1}) and others register_buffer("sqrt_alphas_cumprod", flow.sqrt(alphas_cumprod)) register_buffer("sqrt_one_minus_alphas_cumprod", flow.sqrt(1.0 - alphas_cumprod)) register_buffer("log_one_minus_alphas_cumprod", flow.log(1.0 - alphas_cumprod)) register_buffer("sqrt_recip_alphas_cumprod", flow.sqrt(1.0 / alphas_cumprod)) register_buffer("sqrt_recipm1_alphas_cumprod", flow.sqrt(1.0 / alphas_cumprod - 1)) # calculations for posterior q(x_{t-1} | x_t, x_0) posterior_variance = betas * (1.0 - alphas_cumprod_prev) / (1.0 - alphas_cumprod) # above: equal to 1. / (1. / (1. - alpha_cumprod_tm1) + alpha_t / beta_t) register_buffer("posterior_variance", posterior_variance) register_buffer( "posterior_log_variance_clipped", flow.log(posterior_variance.clamp(min=1e-20)) ) register_buffer( "posterior_mean_coef1", betas * flow.sqrt(alphas_cumprod_prev) / (1.0 - alphas_cumprod) ) register_buffer( "posterior_mean_coef2", (1.0 - alphas_cumprod_prev) * flow.sqrt(alphas) / (1.0 - alphas_cumprod), ) # p2 loss reweighting self.has_p2_loss_reweighting = p2_loss_weight_gamma > 0.0 register_buffer( "p2_loss_weight", (p2_loss_weight_k + alphas_cumprod / (1 - alphas_cumprod)) ** -p2_loss_weight_gamma, ) def q_posterior(self, x_start, x_t, t): posterior_mean = ( extract(self.posterior_mean_coef1, t, x_t.shape) * x_start + extract(self.posterior_mean_coef2, t, x_t.shape) * x_t ) posterior_variance = extract(self.posterior_variance, t, x_t.shape) posterior_log_variance_clipped = extract(self.posterior_log_variance_clipped, t, x_t.shape) return posterior_mean, posterior_variance, posterior_log_variance_clipped def q_sample(self, x_start, t, noise=None): noise = default(noise, lambda: flow.randn_like(x_start)) return ( extract(self.sqrt_alphas_cumprod, t, x_start.shape) * x_start + extract(self.sqrt_one_minus_alphas_cumprod, t, x_start.shape) * noise ) def predict_start_from_noise(self, x_t, t, noise): return ( extract(self.sqrt_recip_alphas_cumprod, t, x_t.shape) * x_t - extract(self.sqrt_recipm1_alphas_cumprod, t, x_t.shape) * noise ) def p2_reweigh_loss(self, loss, times): if not self.has_p2_loss_reweighting: return loss return loss * extract(self.p2_loss_weight, times, loss.shape) # diffusion prior class ChanLayerNorm(nn.Module): def __init__(self, dim, eps=1e-5): super().__init__() self.eps = eps self.weight = nn.Parameter(flow.ones(1, dim, 1, 1)) def forward(self, x): # var = flow.var(x, dim = 1, unbiased = False, keepdim = True) # mean = flow.mean(x, dim = 1, keepdim = True) # return (x - mean) / (var + self.eps).sqrt() * self.weight x = x.permute(0, 2, 3, 1) out = layer_norm(x, normalized_shape=(x.shape[-1:]), eps=self.eps) return out.permute(0, 3, 1, 2) * self.weight class Residual(nn.Module): def __init__(self, fn): super().__init__() self.fn = fn def forward(self, x, **kwargs): return self.fn(x, **kwargs) + x # mlp class MLP(nn.Module): def __init__( self, dim_in, dim_out, *, expansion_factor=2.0, depth=2, norm=False, ): super().__init__() hidden_dim = int(expansion_factor * dim_out) def norm_fn(): return LayerNorm(hidden_dim) if norm else nn.Identity() layers = [nn.Sequential(Linear(dim_in, hidden_dim), nn.SiLU(), norm_fn())] for _ in range(depth - 1): layers.append(nn.Sequential(Linear(hidden_dim, hidden_dim), nn.SiLU(), norm_fn())) layers.append(Linear(hidden_dim, dim_out)) self.net = nn.Sequential(*layers) def forward(self, x): return self.net(x.float()) # relative positional bias for causal transformer class RelPosBias(nn.Module): def __init__( self, heads=8, num_buckets=32, max_distance=128, ): super().__init__() self.num_buckets = num_buckets self.max_distance = max_distance self.relative_attention_bias = Embedding(num_buckets, heads) @staticmethod def _relative_position_bucket(relative_position, num_buckets=32, max_distance=128): n = -relative_position n = flow.max(n, flow.zeros_like(n)).long() max_exact = num_buckets // 2 is_small = n < max_exact val_if_large = ( max_exact + ( flow.log(n.float() / max_exact) / math.log(max_distance / max_exact) * (num_buckets - max_exact) ).long() ) val_if_large = flow.min(val_if_large, flow.zeros_like(val_if_large) + num_buckets - 1) return flow.where(is_small, n, val_if_large) def forward(self, i, j): q_pos = flow.arange(i, dtype=flow.long) k_pos = flow.arange(j, dtype=flow.long) rel_pos = rearrange(k_pos, "j -> 1 j") - rearrange(q_pos, "i -> i 1") rp_bucket = self._relative_position_bucket( rel_pos, num_buckets=self.num_buckets, max_distance=self.max_distance ) values = self.relative_attention_bias( rp_bucket.to_global(sbp=get_default_sbp(), placement=get_default_placement()) ) return rearrange(values, "i j h -> h i j") # feedforward class SwiGLU(nn.Module): """used successfully in https://arxiv.org/abs/2204.0231""" def forward(self, x): x, gate = x.chunk(2, dim=-1) return x * F.silu(gate) def FeedForward(dim, mult=4, dropout=0.0, post_activation_norm=False): """post-activation norm https://arxiv.org/abs/2110.09456""" inner_dim = int(mult * dim) return nn.Sequential( LayerNorm(dim), Linear(dim, inner_dim * 2, bias=False, parallel="col"), SwiGLU(), LayerNorm(inner_dim) if post_activation_norm else nn.Identity(), nn.Dropout(dropout), Linear(inner_dim, dim, bias=False, parallel="row"), ) # attention class Attention(nn.Module): def __init__(self, dim, *, dim_head=64, heads=8, dropout=0.0, causal=False, rotary_emb=None): super().__init__() self.scale = dim_head ** -0.5 self.heads = heads inner_dim = dim_head * heads self.causal = causal self.norm = LayerNorm(dim) self.dropout = nn.Dropout(dropout) self.null_kv = nn.Parameter(flow.randn(2, dim_head)) self.to_q = Linear(dim, inner_dim, bias=False, parallel="col") self.to_kv = Linear(dim, dim_head * 2, bias=False, parallel="col") self.rotary_emb = rotary_emb self.to_out = nn.Sequential( Linear(inner_dim, dim, bias=False, parallel="row"), LayerNorm(dim) ) def forward(self, x, mask=None, attn_bias=None): b, n = x.shape[:2] x = self.norm(x) q, k, v = (self.to_q(x), *self.to_kv(x).chunk(2, dim=-1)) q = rearrange(q, "b n (h d) -> b h n d", h=self.heads) q = q * self.scale # rotary embeddings if exists(self.rotary_emb): q, k = map(self.rotary_emb.rotate_queries_or_keys, (q, k)) nk, nv = repeat_many(self.null_kv.unbind(dim=-2), "d -> b 1 d", b=b) k = flow.cat((nk, k), dim=-2) v = flow.cat((nv, v), dim=-2) # calculate query / key similarities sim = einsum("b h i d, b j d -> b h i j", q, k) # relative positional encoding (T5 style) if exists(attn_bias): sim = sim + attn_bias # masking max_neg_value = -3.4028e38 # flow.finfo(sim.dtype).max if exists(mask): mask = F.pad(mask, (1, 0), value=1) mask = rearrange(mask, "b j -> b 1 1 j") sim = sim.masked_fill(1 - mask, max_neg_value) # ~mask if self.causal: i, j = sim.shape[-2:] causal_mask = flow.ones( (i, j), placement=get_default_placement(), sbp=get_default_sbp(), dtype=flow.int32 ).triu(j - i + 1) sim = sim.masked_fill(causal_mask, max_neg_value) # attention attn = sim.softmax(dim=-1) attn = self.dropout(attn) # aggregate values out = einsum("b h i j, b j d -> b h i d", attn, v) out = rearrange(out, "b h n d -> b n (h d)") return self.to_out(out) class CausalTransformer(nn.Module): def __init__( self, *, dim, depth, dim_head=64, heads=8, ff_mult=4, norm_out=True, attn_dropout=0.0, ff_dropout=0.0, final_proj=True, normformer=False, rotary_emb=True, ): super().__init__() self.rel_pos_bias = RelPosBias(heads=heads) rotary_emb = RotaryEmbedding(dim=min(32, dim_head)) if rotary_emb else None self.layers = nn.ModuleList([]) for _ in range(depth): self.layers.append( nn.ModuleList( [ Attention( dim=dim, causal=True, dim_head=dim_head, heads=heads, dropout=attn_dropout, rotary_emb=rotary_emb, ), FeedForward( dim=dim, mult=ff_mult, dropout=ff_dropout, post_activation_norm=normformer, ), ] ) ) self.norm = LayerNorm(dim) if norm_out else nn.Identity() self.project_out = Linear(dim, dim, bias=False) if final_proj else nn.Identity() def forward( self, x, mask=None, ): n = x.shape[1] attn_bias = self.rel_pos_bias(n, n + 1) for attn, ff in self.layers: x = attn(x, mask=mask, attn_bias=attn_bias) + x x = ff(x) + x out = self.norm(x) return self.project_out(out) class DiffusionPriorNetwork(nn.Module): def __init__( self, dim, num_timesteps=None, num_time_embeds=1, num_image_embeds=1, num_text_embeds=1, max_text_len=256, **kwargs, ): super().__init__() self.num_time_embeds = num_time_embeds self.num_image_embeds = num_image_embeds self.num_text_embeds = num_text_embeds self.to_text_embeds = nn.Sequential( Linear(dim, dim * num_text_embeds) if num_text_embeds > 1 else nn.Identity(), Rearrange("b (n d) -> b n d", n=num_text_embeds), ) self.to_time_embeds = nn.Sequential( Embedding(num_timesteps, dim * num_time_embeds) if exists(num_timesteps) else nn.Sequential( SinusoidalPosEmb(dim), MLP(dim, dim * num_time_embeds) ), # also offer a continuous version of timestep embeddings, with a 2 layer MLP Rearrange("b (n d) -> b n d", n=num_time_embeds), ) self.to_image_embeds = nn.Sequential( Linear(dim, dim * num_image_embeds) if num_image_embeds > 1 else nn.Identity(), Rearrange("b (n d) -> b n d", n=num_image_embeds), ) self.learned_query = nn.Parameter(flow.randn(dim)) self.causal_transformer = CausalTransformer(dim=dim, **kwargs) def forward_with_cond_scale(self, *args, cond_scale=1.0, **kwargs): logits = self.forward(*args, **kwargs) if cond_scale == 1: return logits null_logits = self.forward(*args, cond_drop_prob=1.0, **kwargs) return null_logits + (logits - null_logits) * cond_scale def forward( self, image_embed, diffusion_timesteps, *, text_embed, text_encodings=None, mask=None, cond_drop_prob=0.0, ): batch, dim, dtype = *image_embed.shape, image_embed.dtype num_time_embeds, num_image_embeds, num_text_embeds = ( self.num_time_embeds, self.num_image_embeds, self.num_text_embeds, ) text_embed = self.to_text_embeds(text_embed) image_embed = self.to_image_embeds(image_embed) # make text encodings optional # although the paper seems to suggest it is present <-- if not exists(text_encodings): text_encodings = flow.empty( (batch, 0, dim), placement=get_default_placement(), sbp=get_default_sbp(), dtype=dtype, ) if not exists(mask): mask = flow.ones( (batch, text_encodings.shape[-2]), placement=get_default_placement(), sbp=get_default_sbp(), dtype=flow.bool, ) # classifier free guidance keep_mask = prob_mask_like((batch,), 1 - cond_drop_prob) keep_mask = rearrange(keep_mask, "b -> b 1").to_global( placement=get_default_placement(), sbp=get_default_sbp() ) mask &= keep_mask keep_mask = repeat(keep_mask, "b 1 -> b n", n=num_text_embeds) mask = flow.cat((mask, keep_mask), dim=1) if exists(mask): attend_padding = ( 1 + num_time_embeds + num_image_embeds ) # 1 for learned queries + number of image embeds + time embeds mask = F.pad(mask.to(flow.int32), (0, attend_padding), value=1) time_embed = self.to_time_embeds( diffusion_timesteps.to_global(placement=get_default_placement(), sbp=get_default_sbp()) ) learned_queries = repeat(self.learned_query, "d -> b 1 d", b=batch) tokens = flow.cat( (text_encodings, text_embed, time_embed, image_embed, learned_queries), dim=-2 ) # attend tokens = self.causal_transformer(tokens, mask=mask) # get learned query, which should predict the image embedding (per DDPM timestep) pred_image_embed = tokens[..., -1, :] return pred_image_embed class DiffusionPrior(nn.Module): def __init__( self, net, *, clip=None, image_embed_dim=None, image_size=None, image_channels=3, timesteps=1000, cond_drop_prob=0.0, loss_type="l2", predict_x_start=True, beta_schedule="cosine", condition_on_text_encodings=True, sampling_clamp_l2norm=False, training_clamp_l2norm=False, init_image_embed_l2norm=False, image_embed_scale=None, clip_adapter_overrides=dict(), ): super().__init__() self.noise_scheduler = NoiseScheduler( beta_schedule=beta_schedule, timesteps=timesteps, loss_type=loss_type ) if exists(clip): assert ( image_channels == clip.image_channels ), f"channels of image ({image_channels}) should be equal to the channels " "that CLIP accepts ({clip.image_channels})" freeze_model_and_make_eval_(clip) self.clip = clip else: assert exists( image_embed_dim ), "latent dimension must be given, if training prior network without CLIP given" self.clip = None self.net = net self.image_embed_dim = default(image_embed_dim, lambda: clip.dim_latent) self.channels = default(image_channels, lambda: clip.image_channels) self.cond_drop_prob = cond_drop_prob self.can_classifier_guidance = cond_drop_prob > 0.0 self.condition_on_text_encodings = condition_on_text_encodings self.predict_x_start = predict_x_start self.image_embed_scale = default(image_embed_scale, self.image_embed_dim ** 0.5) # whether to force an l2norm, similar to clipping denoised, when sampling self.sampling_clamp_l2norm = sampling_clamp_l2norm self.training_clamp_l2norm = training_clamp_l2norm self.init_image_embed_l2norm = init_image_embed_l2norm # device tracker self.register_buffer("_dummy", flow.tensor([True]), persistent=False) def p_mean_variance(self, x, t, text_cond, clip_denoised=False, cond_scale=1.0): assert not ( cond_scale != 1.0 and not self.can_classifier_guidance ), "the model was not trained with conditional dropout, " "and thus one cannot use classifier free guidance (cond_scale anything other than 1)" pred = self.net.forward_with_cond_scale(x, t, cond_scale=cond_scale, **text_cond) if self.predict_x_start: x_recon = pred else: x_recon = self.noise_scheduler.predict_start_from_noise(x, t=t, noise=pred) if clip_denoised and not self.predict_x_start: x_recon.clamp_(-1.0, 1.0) if self.predict_x_start and self.sampling_clamp_l2norm: x_recon = l2norm(x_recon) * self.image_embed_scale model_mean, posterior_variance, posterior_log_variance = self.noise_scheduler.q_posterior( x_start=x_recon, x_t=x, t=t ) return model_mean, posterior_variance, posterior_log_variance @flow.no_grad() def p_sample(self, x, t, text_cond=None, clip_denoised=True, cond_scale=1.0): b = x.shape[0] model_mean, _, model_log_variance = self.p_mean_variance( x=x, t=t, text_cond=text_cond, clip_denoised=clip_denoised, cond_scale=cond_scale ) noise = flow.randn(*x.shape, placement=get_default_placement(), sbp=get_default_sbp()) # no noise when t == 0 nonzero_mask = (1 - (t == 0).float()).reshape(b, *((1,) * (len(x.shape) - 1))) return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise @flow.no_grad() def p_sample_loop(self, shape, text_cond, cond_scale=1.0): b = shape[0] image_embed = flow.randn(*shape, placement=get_default_placement(), sbp=get_default_sbp()) if self.init_image_embed_l2norm: image_embed = l2norm(image_embed) * self.image_embed_scale for i in tqdm( reversed(range(0, self.noise_scheduler.num_timesteps)), desc="sampling loop time step", total=self.noise_scheduler.num_timesteps, ): times = flow.full( (b,), i, placement=get_default_placement(), sbp=get_default_sbp(), dtype=flow.long ) image_embed = self.p_sample( image_embed, times, text_cond=text_cond, cond_scale=cond_scale ) return image_embed def p_losses(self, image_embed, times, text_cond, noise=None): noise = default( noise, lambda: flow.randn( *image_embed.shape, placement=get_default_placement(), sbp=get_default_sbp() ), ) image_embed_noisy = self.noise_scheduler.q_sample(x_start=image_embed, t=times, noise=noise) pred = self.net(image_embed_noisy, times, cond_drop_prob=self.cond_drop_prob, **text_cond) if self.predict_x_start and self.training_clamp_l2norm: pred = l2norm(pred) * self.image_embed_scale target = noise if not self.predict_x_start else image_embed loss = self.noise_scheduler.loss_fn(pred, target) return loss @flow.no_grad() @eval_decorator def sample_batch_size(self, batch_size, text_cond, cond_scale=1.0): shape = (batch_size, self.image_embed_dim) img = flow.randn(*shape, placement=get_default_placement(), sbp=get_default_sbp()) for i in tqdm( reversed(range(0, self.noise_scheduler.num_timesteps)), desc="sampling loop time step", total=self.noise_scheduler.num_timesteps, ): img = self.p_sample( img, flow.full( (batch_size,), i, placement=get_default_placement(), sbp=get_default_sbp(), dtype=flow.long, ), text_cond=text_cond, cond_scale=cond_scale, ) return img @flow.no_grad() @eval_decorator def sample( self, text, num_samples_per_batch=2, cond_scale=1.0, text_embed=None, text_encodings=None, text_mask=None, ): # in the paper, what they did was # sample 2 image embeddings, choose the top 1 similarity, as judged by CLIP text = repeat(text, "b ... -> (b r) ...", r=num_samples_per_batch) batch_size = text.shape[0] image_embed_dim = self.image_embed_dim if text_embed is None: assert self.clip is not None text_embed, text_encodings, text_mask = self.clip.embed_text(text) text_cond = dict(text_embed=text_embed) if self.condition_on_text_encodings: text_cond = {**text_cond, "text_encodings": text_encodings, "mask": text_mask} image_embeds = self.p_sample_loop( (batch_size, image_embed_dim), text_cond=text_cond, cond_scale=cond_scale ) # retrieve original unscaled image embed image_embeds /= self.image_embed_scale text_embeds = text_cond["text_embed"] text_embeds = rearrange(text_embeds, "(b r) d -> b r d", r=num_samples_per_batch) image_embeds = rearrange(image_embeds, "(b r) d -> b r d", r=num_samples_per_batch) text_image_sims = einsum("b r d, b r d -> b r", l2norm(text_embeds), l2norm(image_embeds)) top_sim_indices = text_image_sims.topk(k=1)[1] # .indices top_sim_indices = repeat(top_sim_indices, "b 1 -> b 1 d", d=image_embed_dim) top_image_embeds = image_embeds.gather(1, top_sim_indices) return rearrange(top_image_embeds, "b 1 d -> b d") def forward( self, text=None, image=None, text_embed=None, # allow for training on preprocessed CLIP text and image embeddings image_embed=None, text_encodings=None, # as well as CLIP text encodings text_mask=None, *args, **kwargs, ): assert exists(text) ^ exists(text_embed), "either text or text embedding must be supplied" assert exists(image) ^ exists(image_embed), "either text or text embedding must be supplied" assert not ( self.condition_on_text_encodings and (not exists(text_encodings) and not exists(text)) ), "text encodings must be present if you specified to condition on it on initialization" if exists(image): image_embed, _ = self.clip.embed_image(image) # calculate text conditionings, based on what is passed in if exists(text): text_embed, text_encodings, text_mask = self.clip.embed_text(text) text_cond = dict(text_embed=text_embed) if self.condition_on_text_encodings: assert exists( text_encodings ), "text encodings must be present for diffusion prior if specified" text_cond = {**text_cond, "text_encodings": text_encodings, "mask": text_mask} # timestep conditioning from ddpm batch = image_embed.shape[0] times = flow.randint( 0, self.noise_scheduler.num_timesteps, (batch,), placement=get_default_placement(), sbp=get_default_sbp(), dtype=flow.long, ) # scale image embed (Katherine) image_embed *= self.image_embed_scale # calculate forward loss return self.p_losses(image_embed, times, text_cond=text_cond, *args, **kwargs) # decoder def ConvTransposeUpsample(dim, dim_out=None): dim_out = default(dim_out, dim) return ConvTranspose2d(dim, dim_out, 4, 2, 1) def NearestUpsample(dim, dim_out=None): dim_out = default(dim_out, dim) return nn.Sequential( nn.Upsample(scale_factor=2, mode="nearest"), Conv2d(dim, dim_out, 3, padding=1) ) def Downsample(dim, *, dim_out=None): dim_out = default(dim_out, dim) return Conv2d(dim, dim_out, 4, 2, 1) class SinusoidalPosEmb(nn.Module): def __init__(self, dim): super().__init__() self.dim = dim def forward(self, x): half_dim = self.dim // 2 emb = math.log(10000) / (half_dim - 1) emb = flow.exp( flow.arange(half_dim, placement=get_default_placement(), sbp=get_default_sbp()) * -emb ) emb = rearrange(x, "i -> i 1") * rearrange(emb, "j -> 1 j") return flow.cat((emb.sin(), emb.cos()), dim=-1) class Block(nn.Module): def __init__(self, dim, dim_out, groups=8): super().__init__() self.project = Conv2d(dim, dim_out, 3, padding=1) self.norm = GroupNorm(groups, dim_out) self.act = nn.SiLU() def forward(self, x, scale_shift=None): x = self.project(x) x = self.norm(x) if exists(scale_shift): scale, shift = scale_shift x = x * (scale + 1) + shift x = self.act(x) return x class ResnetBlock(nn.Module): def __init__(self, dim, dim_out, *, cond_dim=None, time_cond_dim=None, groups=8): super().__init__() self.time_mlp = None if exists(time_cond_dim): self.time_mlp = nn.Sequential(nn.SiLU(), Linear(time_cond_dim, dim_out * 2)) self.cross_attn = None if exists(cond_dim): self.cross_attn = EinopsToAndFrom( "b c h w", "b (h w) c", CrossAttention(dim=dim_out, context_dim=cond_dim) ) self.block1 = Block(dim, dim_out, groups=groups) self.block2 = Block(dim_out, dim_out, groups=groups) self.res_conv = Conv2d(dim, dim_out, 1) if dim != dim_out else nn.Identity() def forward(self, x, time_emb=None, cond=None): scale_shift = None if exists(self.time_mlp) and exists(time_emb): time_emb = self.time_mlp(time_emb) time_emb = rearrange(time_emb, "b c -> b c 1 1") scale_shift = time_emb.chunk(2, dim=1) h = self.block1(x, scale_shift=scale_shift) if exists(self.cross_attn): assert exists(cond) h = self.cross_attn(h, context=cond) + h h = self.block2(h) return h + self.res_conv(x) class CrossAttention(nn.Module): def __init__( self, dim, *, context_dim=None, dim_head=64, heads=8, dropout=0.0, norm_context=False ): super().__init__() self.scale = dim_head ** -0.5 self.heads = heads inner_dim = dim_head * heads context_dim = default(context_dim, dim) self.norm = LayerNorm(dim) self.norm_context = LayerNorm(context_dim) if norm_context else nn.Identity() self.dropout = nn.Dropout(dropout) self.null_kv = nn.Parameter( flow.randn(2, dim_head, placement=get_default_placement(), sbp=get_default_sbp()) ) self.to_q = Linear(dim, inner_dim, bias=False, parallel="col") self.to_kv = Linear(context_dim, inner_dim * 2, bias=False, parallel="col") self.to_out = nn.Sequential( Linear(inner_dim, dim, bias=False, parallel="row"), LayerNorm(dim) ) def forward(self, x, context, mask=None): b, n = x.shape[:2] x = self.norm(x) context = self.norm_context(context) q, k, v = (self.to_q(x), *self.to_kv(context).chunk(2, dim=-1)) q, k, v = rearrange_many((q, k, v), "b n (h d) -> b h n d", h=self.heads) # add null key / value for classifier free guidance in prior net nk, nv = repeat_many(self.null_kv.unbind(dim=-2), "d -> b h 1 d", h=self.heads, b=b) k = flow.cat((nk, k), dim=-2) v = flow.cat((nv, v), dim=-2) q = q * self.scale sim = einsum("b h i d, b h j d -> b h i j", q, k) max_neg_value = -3.4028e38 # -flow.finfo(sim.dtype).max if exists(mask): mask = F.pad(mask, (1, 0), value=1) mask = rearrange(mask, "b j -> b 1 1 j") sim = sim.masked_fill(1 - mask, max_neg_value) attn = sim.softmax(dim=-1) out = einsum("b h i j, b h j d -> b h i d", attn, v) out = rearrange(out, "b h n d -> b n (h d)") return self.to_out(out) class LinearAttention(nn.Module): def __init__(self, dim, dim_head=32, heads=8): super().__init__() self.scale = dim_head ** -0.5 self.heads = heads inner_dim = dim_head * heads self.norm = ChanLayerNorm(dim) self.nonlin = nn.GELU() self.to_qkv = Conv2d(dim, inner_dim * 3, 1, bias=False) self.to_out = nn.Sequential(Conv2d(inner_dim, dim, 1, bias=False), ChanLayerNorm(dim)) def forward(self, fmap): h, x, y = self.heads, *fmap.shape[-2:] fmap = self.norm(fmap) q, k, v = self.to_qkv(fmap).chunk(3, dim=1) q, k, v = rearrange_many((q, k, v), "b (h c) x y -> (b h) (x y) c", h=h) q = q.softmax(dim=-1) k = k.softmax(dim=-2) q = q * self.scale context = einsum("b n d, b n e -> b d e", k, v) out = einsum("b n d, b d e -> b n e", q, context) out = rearrange(out, "(b h) (x y) d -> b (h d) x y", h=h, x=x, y=y) out = self.nonlin(out) return self.to_out(out) class CrossEmbedLayer(nn.Module): def __init__(self, dim_in, kernel_sizes, dim_out=None, stride=2): super().__init__() assert all([*map(lambda t: (t % 2) == (stride % 2), kernel_sizes)]) dim_out = default(dim_out, dim_in) kernel_sizes = sorted(kernel_sizes) num_scales = len(kernel_sizes) # calculate the dimension at each scale dim_scales = [int(dim_out / (2 ** i)) for i in range(1, num_scales)] dim_scales = [*dim_scales, dim_out - sum(dim_scales)] self.convs = nn.ModuleList([]) for kernel, dim_scale in zip(kernel_sizes, dim_scales): self.convs.append( Conv2d(dim_in, dim_scale, kernel, stride=stride, padding=(kernel - stride) // 2) ) def forward(self, x): fmaps = tuple(map(lambda conv: conv(x), self.convs)) return flow.cat(fmaps, dim=1) class Unet(nn.Module): def __init__( self, dim, *, image_embed_dim=None, text_embed_dim=None, cond_dim=None, num_image_tokens=4, num_time_tokens=2, out_dim=None, dim_mults=(1, 2, 4, 8), channels=3, channels_out=None, self_attn=False, attn_dim_head=32, attn_heads=16, lowres_cond=False, # sparse_attn=False, attend_at_middle=True, cond_on_text_encodings=False, max_text_len=256, cond_on_image_embeds=False, add_image_embeds_to_time=True, # init_dim=None, init_conv_kernel_size=7, resnet_groups=8, num_resnet_blocks=2, init_cross_embed_kernel_sizes=(3, 7, 15), cross_embed_downsample=False, cross_embed_downsample_kernel_sizes=(2, 4), memory_efficient=False, scale_skip_connection=False, nearest_upsample=False, final_conv_kernel_size=1, **kwargs, ): super().__init__() # save locals to take care of some hyperparameters for cascading DDPM self._locals = locals() del self._locals["self"] del self._locals["__class__"] # for eventual cascading diffusion self.lowres_cond = lowres_cond # determine dimensions self.channels = channels self.channels_out = default(channels_out, channels) init_channels = channels if not lowres_cond else channels * 2 init_dim = default(init_dim, dim) self.init_conv = CrossEmbedLayer( init_channels, dim_out=init_dim, kernel_sizes=init_cross_embed_kernel_sizes, stride=1 ) dims = [init_dim, *map(lambda m: dim * m, dim_mults)] in_out = list(zip(dims[:-1], dims[1:])) num_stages = len(in_out) # time, image embeddings, and optional text encoding cond_dim = default(cond_dim, dim) time_cond_dim = dim * 4 self.to_time_hiddens = nn.Sequential( SinusoidalPosEmb(dim), Linear(dim, time_cond_dim, parallel="col"), nn.GELU() ) self.to_time_tokens = nn.Sequential( Linear(time_cond_dim, cond_dim * num_time_tokens, parallel="row"), Rearrange("b (r d) -> b r d", r=num_time_tokens), ) self.to_time_cond = nn.Sequential(Linear(time_cond_dim, time_cond_dim, parallel="row")) self.image_to_tokens = ( nn.Sequential( Linear(image_embed_dim, cond_dim * num_image_tokens), Rearrange("b (n d) -> b n d", n=num_image_tokens), ) if cond_on_image_embeds and image_embed_dim != cond_dim else nn.Identity() ) self.to_image_hiddens = ( nn.Sequential(Linear(image_embed_dim, time_cond_dim), nn.GELU()) if cond_on_image_embeds and add_image_embeds_to_time else None ) self.norm_cond = LayerNorm(cond_dim) self.norm_mid_cond = LayerNorm(cond_dim) # text encoding conditioning (optional) self.text_to_cond = None if cond_on_text_encodings: assert exists( text_embed_dim ), "text_embed_dim must be given to the unet if cond_on_text_encodings is True" self.text_to_cond = Linear(text_embed_dim, cond_dim) # finer control over whether to condition on image embeddings and text encodings # so one can have the latter unets in the cascading DDPMs only focus on super-resoluting self.cond_on_text_encodings = cond_on_text_encodings self.cond_on_image_embeds = cond_on_image_embeds # for classifier free guidance self.null_image_embed = nn.Parameter(flow.randn(1, num_image_tokens, cond_dim)) self.null_image_hiddens = nn.Parameter(flow.randn(1, time_cond_dim)) self.max_text_len = max_text_len self.null_text_embed = nn.Parameter(flow.randn(1, max_text_len, cond_dim)) # whether to scale skip connection, adopted in Imagen self.skip_connect_scale = 1.0 if not scale_skip_connection else (2 ** -0.5) # attention related params attn_kwargs = dict(heads=attn_heads, dim_head=attn_dim_head) self_attn = cast_tuple(self_attn, num_stages) def create_self_attn(dim): return EinopsToAndFrom("b c h w", "b (h w) c", Residual(Attention(dim, **attn_kwargs))) # resnet block klass resnet_groups = cast_tuple(resnet_groups, num_stages) top_level_resnet_group = first(resnet_groups) num_resnet_blocks = cast_tuple(num_resnet_blocks, num_stages) # downsample klass downsample_klass = Downsample if cross_embed_downsample: downsample_klass = partial( CrossEmbedLayer, kernel_sizes=cross_embed_downsample_kernel_sizes ) # upsample klass upsample_klass = ConvTransposeUpsample if not nearest_upsample else NearestUpsample # give memory efficient unet an initial resnet block self.init_resnet_block = ( ResnetBlock( init_dim, init_dim, time_cond_dim=time_cond_dim, groups=top_level_resnet_group ) if memory_efficient else None ) # layers self.downs = nn.ModuleList([]) self.ups = nn.ModuleList([]) num_resolutions = len(in_out) skip_connect_dims = [] # keeping track of skip connection dimensions for ind, ((dim_in, dim_out), groups, layer_num_resnet_blocks, layer_self_attn) in enumerate( zip(in_out, resnet_groups, num_resnet_blocks, self_attn) ): is_first = ind == 0 is_last = ind >= (num_resolutions - 1) layer_cond_dim = cond_dim if not is_first else None dim_layer = dim_out if memory_efficient else dim_in skip_connect_dims.append(dim_layer) attention = nn.Identity() if layer_self_attn: attention = create_self_attn(dim_layer) elif sparse_attn: attention = Residual(LinearAttention(dim_layer, **attn_kwargs)) self.downs.append( nn.ModuleList( [ downsample_klass(dim_in, dim_out=dim_out) if memory_efficient else None, ResnetBlock( dim_layer, dim_layer, time_cond_dim=time_cond_dim, groups=groups ), nn.ModuleList( [ ResnetBlock( dim_layer, dim_layer, cond_dim=layer_cond_dim, time_cond_dim=time_cond_dim, groups=groups, ) for _ in range(layer_num_resnet_blocks) ] ), attention, downsample_klass(dim_layer, dim_out=dim_out) if not is_last and not memory_efficient else Conv2d(dim_layer, dim_out, 1), ] ) ) mid_dim = dims[-1] self.mid_block1 = ResnetBlock( mid_dim, mid_dim, cond_dim=cond_dim, time_cond_dim=time_cond_dim, groups=resnet_groups[-1], ) self.mid_attn = create_self_attn(mid_dim) self.mid_block2 = ResnetBlock( mid_dim, mid_dim, cond_dim=cond_dim, time_cond_dim=time_cond_dim, groups=resnet_groups[-1], ) for ind, ((dim_in, dim_out), groups, layer_num_resnet_blocks, layer_self_attn) in enumerate( zip( reversed(in_out), reversed(resnet_groups), reversed(num_resnet_blocks), reversed(self_attn), ) ): is_last = ind >= (len(in_out) - 1) layer_cond_dim = cond_dim if not is_last else None skip_connect_dim = skip_connect_dims.pop() attention = nn.Identity() if layer_self_attn: attention = create_self_attn(dim_out) elif sparse_attn: attention = Residual(LinearAttention(dim_out, **attn_kwargs)) self.ups.append( nn.ModuleList( [ ResnetBlock( dim_out + skip_connect_dim, dim_out, cond_dim=layer_cond_dim, time_cond_dim=time_cond_dim, groups=groups, ), nn.ModuleList( [ ResnetBlock( dim_out + skip_connect_dim, dim_out, cond_dim=layer_cond_dim, time_cond_dim=time_cond_dim, groups=groups, ) for _ in range(layer_num_resnet_blocks) ] ), attention, upsample_klass(dim_out, dim_in) if not is_last or memory_efficient else nn.Identity(), ] ) ) self.final_resnet_block = ResnetBlock( dim * 2, dim, time_cond_dim=time_cond_dim, groups=top_level_resnet_group ) self.to_out = Conv2d( dim, self.channels_out, kernel_size=final_conv_kernel_size, padding=final_conv_kernel_size // 2, ) # if the current settings for the unet are not correct # for cascading DDPM, then reinit the unet with the right settings def cast_model_parameters( self, *, lowres_cond, channels, channels_out, cond_on_image_embeds, cond_on_text_encodings ): if ( lowres_cond == self.lowres_cond and channels == self.channels and cond_on_image_embeds == self.cond_on_image_embeds and cond_on_text_encodings == self.cond_on_text_encodings and channels_out == self.channels_out ): return self updated_kwargs = dict( lowres_cond=lowres_cond, channels=channels, channels_out=channels_out, cond_on_image_embeds=cond_on_image_embeds, cond_on_text_encodings=cond_on_text_encodings, ) return self.__class__(**{**self._locals, **updated_kwargs}) def forward_with_cond_scale(self, *args, cond_scale=1.0, **kwargs): logits = self.forward(*args, **kwargs) if cond_scale == 1: return logits null_logits = self.forward( *args, text_cond_drop_prob=1.0, image_cond_drop_prob=1.0, **kwargs ) return null_logits + (logits - null_logits) * cond_scale def forward( self, x, time, *, image_embed, lowres_cond_img=None, text_encodings=None, text_mask=None, image_cond_drop_prob=0.0, text_cond_drop_prob=0.0, blur_sigma=None, blur_kernel_size=None, ): batch_size = x.shape[0] # add low resolution conditioning, if present assert not ( self.lowres_cond and not exists(lowres_cond_img) ), "low resolution conditioning image must be present" if exists(lowres_cond_img): x = flow.cat((x, lowres_cond_img), dim=1) # initial convolution x = self.init_conv(x) r = x.clone() # final residual # time conditioning time_hiddens = self.to_time_hiddens(time) time_tokens = self.to_time_tokens(time_hiddens) t = self.to_time_cond(time_hiddens) # conditional dropout image_keep_mask = prob_mask_like((batch_size,), 1 - image_cond_drop_prob) text_keep_mask = prob_mask_like((batch_size,), 1 - text_cond_drop_prob) text_keep_mask = rearrange(text_keep_mask, "b -> b 1 1") # image embedding to be summed to time embedding # discovered by @mhh0318 in the paper if exists(image_embed) and exists(self.to_image_hiddens): image_hiddens = self.to_image_hiddens(image_embed) image_keep_mask_hidden = rearrange(image_keep_mask, "b -> b 1") null_image_hiddens = self.null_image_hiddens.to(image_hiddens.dtype) image_hiddens = flow.where(image_keep_mask_hidden, image_hiddens, null_image_hiddens) t = t + image_hiddens # mask out image embedding depending on condition dropout # for classifier free guidance image_tokens = None if self.cond_on_image_embeds: image_keep_mask_embed = rearrange(image_keep_mask, "b -> b 1 1") image_tokens = self.image_to_tokens(image_embed) null_image_embed = self.null_image_embed.to( image_tokens.dtype ) # for some reason pyflow AMP not working image_tokens = flow.where(image_keep_mask_embed, image_tokens, null_image_embed) # take care of text encodings (optional) text_tokens = None if exists(text_encodings) and self.cond_on_text_encodings: text_tokens = self.text_to_cond(text_encodings) text_tokens = text_tokens[:, : self.max_text_len] text_tokens_len = text_tokens.shape[1] remainder = self.max_text_len - text_tokens_len if remainder > 0: text_tokens = F.pad(text_tokens, (0, 0, 0, remainder)) if exists(text_mask): if remainder > 0: # text_mask = F.pad(text_mask, (0, remainder), value = False) text_mask = F.pad(text_mask.to(flow.int32), (0, remainder), value=0) text_mask = rearrange(text_mask, "b n -> b n 1") text_keep_mask = text_mask & text_keep_mask null_text_embed = self.null_text_embed.to( text_tokens.dtype ) # for some reason pyflow AMP not working text_tokens = flow.where(text_keep_mask, text_tokens, null_text_embed) # main conditioning tokens (c) c = time_tokens if exists(image_tokens): c = flow.cat((c, image_tokens), dim=-2) # text and image conditioning tokens (mid_c), to save on compute, # only do cross attention based conditioning on the inner most layers of the Unet mid_c = c if not exists(text_tokens) else flow.cat((c, text_tokens), dim=-2) # normalize conditioning tokens c = self.norm_cond(c) mid_c = self.norm_mid_cond(mid_c) # initial resnet block if exists(self.init_resnet_block): x = self.init_resnet_block(x, t) # go through the layers of the unet, down and up hiddens = [] for pre_downsample, init_block, resnet_blocks, attn, post_downsample in self.downs: if exists(pre_downsample): x = pre_downsample(x) x = init_block(x, t, c) for resnet_block in resnet_blocks: x = resnet_block(x, t, c) hiddens.append(x) x = attn(x) hiddens.append(x) if exists(post_downsample): x = post_downsample(x) x = self.mid_block1(x, t, mid_c) if exists(self.mid_attn): x = self.mid_attn(x) x = self.mid_block2(x, t, mid_c) def connect_skip(fmap): return flow.cat((fmap, hiddens.pop() * self.skip_connect_scale), dim=1) for init_block, resnet_blocks, attn, upsample in self.ups: x = connect_skip(x) x = init_block(x, t, c) for resnet_block in resnet_blocks: x = connect_skip(x) x = resnet_block(x, t, c) x = attn(x) x = upsample(x) x = flow.cat((x, r), dim=1) x = self.final_resnet_block(x, t) return self.to_out(x) class LowresConditioner(nn.Module): def __init__( self, downsample_first=True, blur_sigma=0.6, blur_kernel_size=3, ): super().__init__() self.downsample_first = downsample_first self.blur_sigma = blur_sigma self.blur_kernel_size = blur_kernel_size def forward( self, cond_fmap, *, target_image_size, downsample_image_size=None, blur_sigma=None, blur_kernel_size=None, ): if self.training and self.downsample_first and exists(downsample_image_size): cond_fmap = resize_image_to(cond_fmap, downsample_image_size) if self.training: # when training, blur the low resolution conditional image blur_sigma = default(blur_sigma, self.blur_sigma) blur_kernel_size = default(blur_kernel_size, self.blur_kernel_size) # allow for drawing a random sigma between lo and hi float values if isinstance(blur_sigma, tuple): blur_sigma = tuple(map(float, blur_sigma)) blur_sigma = random.uniform(*blur_sigma) # allow for drawing a random kernel size between lo and hi int values if isinstance(blur_kernel_size, tuple): blur_kernel_size = tuple(map(int, blur_kernel_size)) kernel_size_lo, kernel_size_hi = blur_kernel_size blur_kernel_size = random.randrange(kernel_size_lo, kernel_size_hi + 1) cond_fmap = gaussian_blur2d( cond_fmap, cast_tuple(blur_kernel_size, 2), cast_tuple(blur_sigma, 2) ) cond_fmap = resize_image_to(cond_fmap, target_image_size) return cond_fmap class Decoder(nn.Module): def __init__( self, unet, *, clip=None, image_size=None, channels=3, vae=tuple(), timesteps=1000, image_cond_drop_prob=0.1, text_cond_drop_prob=0.5, loss_type="l2", beta_schedule=None, predict_x_start=False, predict_x_start_for_latent_diffusion=False, image_sizes=None, # for cascading ddpm, image size at each stage random_crop_sizes=None, lowres_downsample_first=True, blur_sigma=0.6, # cascading ddpm - blur sigma blur_kernel_size=3, # cascading ddpm - blur kernel size clip_denoised=True, clip_x_start=True, clip_adapter_overrides=dict(), learned_variance=True, learned_variance_constrain_frac=False, vb_loss_weight=0.001, unconditional=False, # set to True for generating images without conditioning auto_normalize_img=True, use_dynamic_thres=False, # from the Imagen paper dynamic_thres_percentile=0.9, p2_loss_weight_gamma=0.0, p2_loss_weight_k=1, ): super().__init__() # clip self.clip = None if exists(clip): assert not unconditional, "clip must not be given if doing unconditional image training" assert ( channels == clip.image_channels ), f"channels of image ({channels}) should be equal to the" " channels that CLIP accepts ({clip.image_channels})" freeze_model_and_make_eval_(clip) self.clip = clip # determine image size, with image_size and image_sizes taking precedence if exists(image_size) or exists(image_sizes): assert exists(image_size) ^ exists( image_sizes ), "only one of image_size or image_sizes must be given" image_size = default(image_size, lambda: image_sizes[-1]) elif exists(clip): image_size = clip.image_size else: raise ("either image_size, image_sizes, or clip must be given to decoder") # channels self.channels = channels # verify conditioning method unets = cast_tuple(unet) num_unets = len(unets) self.unconditional = unconditional # automatically take care of ensuring that first unet is unconditional while the rest # of the unets are conditioned on the low resolution image produced by previous unet vaes = pad_tuple_to_length( cast_tuple(vae), len(unets), fillvalue=NullVQGanVAE(channels=self.channels) ) # whether to use learned variance, defaults to True for the first unet in the cascade learned_variance = pad_tuple_to_length( cast_tuple(learned_variance), len(unets), fillvalue=False ) self.learned_variance = learned_variance # whether to constrain the output of the network (the interpolation fraction) from 0 to 1 self.learned_variance_constrain_frac = learned_variance_constrain_frac self.vb_loss_weight = vb_loss_weight # construct unets and vaes self.unets = nn.ModuleList([]) self.vaes = nn.ModuleList([]) for ind, (one_unet, one_vae, one_unet_learned_var) in enumerate( zip(unets, vaes, learned_variance) ): assert isinstance(one_unet, Unet) assert isinstance(one_vae, (VQGanVAE, NullVQGanVAE)) is_first = ind == 0 latent_dim = one_vae.encoded_dim if exists(one_vae) else None unet_channels = default(latent_dim, self.channels) unet_channels_out = unet_channels * (1 if not one_unet_learned_var else 2) one_unet = one_unet.cast_model_parameters( lowres_cond=not is_first, cond_on_image_embeds=not unconditional and is_first, cond_on_text_encodings=not unconditional and one_unet.cond_on_text_encodings, channels=unet_channels, channels_out=unet_channels_out, ) self.unets.append(one_unet) self.vaes.append(one_vae.copy_for_eval()) # determine from unets whether conditioning on text encoding is needed self.condition_on_text_encodings = any([unet.cond_on_text_encodings for unet in self.unets]) # create noise schedulers per unet if not exists(beta_schedule): beta_schedule = ( "cosine", *(("cosine",) * max(num_unets - 2, 0)), *(("linear",) * int(num_unets > 1)), ) beta_schedule = cast_tuple(beta_schedule, num_unets) p2_loss_weight_gamma = cast_tuple(p2_loss_weight_gamma, num_unets) self.noise_schedulers = nn.ModuleList([]) for unet_beta_schedule, unet_p2_loss_weight_gamma in zip( beta_schedule, p2_loss_weight_gamma ): noise_scheduler = NoiseScheduler( beta_schedule=unet_beta_schedule, timesteps=timesteps, loss_type=loss_type, p2_loss_weight_gamma=unet_p2_loss_weight_gamma, p2_loss_weight_k=p2_loss_weight_k, ) self.noise_schedulers.append(noise_scheduler) # unet image sizes image_sizes = default(image_sizes, (image_size,)) image_sizes = tuple(sorted(set(image_sizes))) assert len(self.unets) == len( image_sizes ), "you did not supply the correct number of u-nets " f"({len(self.unets)}) for resolutions {image_sizes}" self.image_sizes = image_sizes self.sample_channels = cast_tuple(self.channels, len(image_sizes)) # random crop sizes (for super-resoluting unets at the end of cascade?) self.random_crop_sizes = cast_tuple(random_crop_sizes, len(image_sizes)) # predict x0 config self.predict_x_start = ( cast_tuple(predict_x_start, len(unets)) if not predict_x_start_for_latent_diffusion else tuple(map(lambda t: isinstance(t, VQGanVAE), self.vaes)) ) # cascading ddpm related stuff lowres_conditions = tuple(map(lambda t: t.lowres_cond, self.unets)) assert lowres_conditions == ( False, *((True,) * (len(self.unets) - 1)), ), "the first unet must be unconditioned (by low resolution image), " "and the rest of the unets must have `lowres_cond` set to True" self.to_lowres_cond = LowresConditioner( downsample_first=lowres_downsample_first, blur_sigma=blur_sigma, blur_kernel_size=blur_kernel_size, ) # classifier free guidance self.image_cond_drop_prob = image_cond_drop_prob self.text_cond_drop_prob = text_cond_drop_prob self.can_classifier_guidance = image_cond_drop_prob > 0.0 or text_cond_drop_prob > 0.0 # whether to clip when sampling self.clip_denoised = clip_denoised self.clip_x_start = clip_x_start # dynamic thresholding settings, if clipping denoised during sampling self.use_dynamic_thres = use_dynamic_thres self.dynamic_thres_percentile = dynamic_thres_percentile # normalize and unnormalize image functions self.normalize_img = normalize_neg_one_to_one if auto_normalize_img else identity self.unnormalize_img = unnormalize_zero_to_one if auto_normalize_img else identity # device tracker self.register_buffer("_dummy", flow.Tensor([True]), persistent=False) def get_unet(self, unet_number): assert 0 < unet_number <= len(self.unets) index = unet_number - 1 return self.unets[index] @contextmanager def one_unet_in_gpu(self, unet_number=None, unet=None): assert exists(unet_number) ^ exists(unet) if exists(unet_number): unet = self.get_unet(unet_number) self.cuda() devices = [module_device(unet) for unet in self.unets] self.unets.cpu() unet.cuda() yield for unet, device in zip(self.unets, devices): unet.to(device) def p_mean_variance( self, unet, x, t, image_embed, noise_scheduler, text_encodings=None, text_mask=None, lowres_cond_img=None, clip_denoised=True, predict_x_start=False, learned_variance=False, cond_scale=1.0, model_output=None, ): assert not ( cond_scale != 1.0 and not self.can_classifier_guidance ), "the decoder was not trained with conditional dropout, " "and thus one cannot use classifier free guidance (cond_scale anything other than 1)" pred = default( model_output, lambda: unet.forward_with_cond_scale( x, t, image_embed=image_embed, text_encodings=text_encodings, text_mask=text_mask, cond_scale=cond_scale, lowres_cond_img=lowres_cond_img, ), ) if learned_variance: pred, var_interp_frac_unnormalized = pred.chunk(2, dim=1) if predict_x_start: x_recon = pred else: x_recon = noise_scheduler.predict_start_from_noise(x, t=t, noise=pred) if clip_denoised: # s is the threshold amount # static thresholding would just be s = 1 s = 1.0 if self.use_dynamic_thres: s = flow.quantile( rearrange(x_recon, "b ... -> b (...)").abs(), self.dynamic_thres_percentile, dim=-1, ) s.clamp_(min=1.0) s = s.view(-1, *((1,) * (x_recon.ndim - 1))) # clip by threshold, depending on whether static or dynamic x_recon = x_recon.clamp(-s, s) / s model_mean, posterior_variance, posterior_log_variance = noise_scheduler.q_posterior( x_start=x_recon, x_t=x, t=t ) if learned_variance: # if learned variance, posterio variance and posterior log variance are # predicted by the network by an interpolation of the max and min log beta values # eq 15 - https://arxiv.org/abs/2102.09672 min_log = extract(noise_scheduler.posterior_log_variance_clipped, t, x.shape) max_log = extract(flow.log(noise_scheduler.betas), t, x.shape) var_interp_frac = unnormalize_zero_to_one(var_interp_frac_unnormalized) if self.learned_variance_constrain_frac: var_interp_frac = var_interp_frac.sigmoid() posterior_log_variance = var_interp_frac * max_log + (1 - var_interp_frac) * min_log posterior_variance = posterior_log_variance.exp() return model_mean, posterior_variance, posterior_log_variance @flow.no_grad() def p_sample( self, unet, x, t, image_embed, noise_scheduler, text_encodings=None, text_mask=None, cond_scale=1.0, lowres_cond_img=None, predict_x_start=False, learned_variance=False, clip_denoised=True, ): b = x.shape[0] model_mean, _, model_log_variance = self.p_mean_variance( unet, x=x, t=t, image_embed=image_embed, text_encodings=text_encodings, text_mask=text_mask, cond_scale=cond_scale, lowres_cond_img=lowres_cond_img, clip_denoised=clip_denoised, predict_x_start=predict_x_start, noise_scheduler=noise_scheduler, learned_variance=learned_variance, ) noise = flow.randn(*x.shape, placement=get_default_placement(), sbp=get_default_sbp()) # no noise when t == 0 nonzero_mask = (1 - (t == 0).float()).reshape(b, *((1,) * (len(x.shape) - 1))) return model_mean + nonzero_mask * (0.5 * model_log_variance).exp() * noise @flow.no_grad() def p_sample_loop( self, unet, shape, image_embed, noise_scheduler, predict_x_start=False, learned_variance=False, clip_denoised=True, lowres_cond_img=None, text_encodings=None, text_mask=None, cond_scale=1, is_latent_diffusion=False, ): b = shape[0] img = flow.randn(*shape, placement=get_default_placement(), sbp=get_default_sbp()) if not is_latent_diffusion: lowres_cond_img = maybe(self.normalize_img)(lowres_cond_img) for i in tqdm( reversed(range(0, noise_scheduler.num_timesteps)), desc="sampling loop time step", total=noise_scheduler.num_timesteps, ): img = self.p_sample( unet, img, flow.full( (b,), i, placement=get_default_placement(), sbp=get_default_sbp(), dtype=flow.long, ), image_embed=image_embed, text_encodings=text_encodings, text_mask=text_mask, cond_scale=cond_scale, lowres_cond_img=lowres_cond_img, predict_x_start=predict_x_start, noise_scheduler=noise_scheduler, learned_variance=learned_variance, clip_denoised=clip_denoised, ) unnormalize_img = self.unnormalize_img(img) return unnormalize_img def p_losses( self, unet, x_start, times, *, image_embed, noise_scheduler, lowres_cond_img=None, text_encodings=None, text_mask=None, predict_x_start=False, noise=None, learned_variance=False, clip_denoised=False, is_latent_diffusion=False, ): noise = default( noise, lambda: flow.randn( *x_start.shape, placement=get_default_placement(), sbp=get_default_sbp() ), ) # normalize to [-1, 1] if not is_latent_diffusion: x_start = self.normalize_img(x_start) lowres_cond_img = maybe(self.normalize_img)(lowres_cond_img) # get x_t x_noisy = noise_scheduler.q_sample(x_start=x_start, t=times, noise=noise) model_output = unet( x_noisy, times, image_embed=image_embed, text_encodings=text_encodings, text_mask=text_mask, lowres_cond_img=lowres_cond_img, image_cond_drop_prob=self.image_cond_drop_prob, text_cond_drop_prob=self.text_cond_drop_prob, ) if learned_variance: pred, _ = model_output.chunk(2, dim=1) else: pred = model_output target = noise if not predict_x_start else x_start loss = noise_scheduler.loss_fn(pred, target, reduction="none") loss = reduce(loss, "b ... -> b (...)", "mean") loss = noise_scheduler.p2_reweigh_loss(loss, times) loss = loss.mean() if not learned_variance: # return simple loss if not using learned variance return loss # most of the code below is transcribed from # https://github.com/hojonathanho/diffusion/blob/master/diffusion_tf/diffusion_utils_2.py # the Improved DDPM paper then further modified it so that the mean is detached # (shown a couple lines before), and weighted to be smaller than the l1 or l2 "simple" loss # it is questionable whether this is really needed, looking at some of the figures in the # paper, but may as well stay faithful to their implementation # if learning the variance, also include the extra weight kl loss true_mean, _, true_log_variance_clipped = noise_scheduler.q_posterior( x_start=x_start, x_t=x_noisy, t=times ) model_mean, _, model_log_variance = self.p_mean_variance( unet, x=x_noisy, t=times, image_embed=image_embed, noise_scheduler=noise_scheduler, clip_denoised=clip_denoised, learned_variance=True, model_output=model_output, ) # kl loss with detached model predicted mean, for stability reasons as in paper detached_model_mean = model_mean.detach() kl = normal_kl( true_mean, true_log_variance_clipped, detached_model_mean, model_log_variance ) kl = meanflat(kl) * NAT decoder_nll = -discretized_gaussian_log_likelihood( x_start, means=detached_model_mean, log_scales=0.5 * model_log_variance ) decoder_nll = meanflat(decoder_nll) * NAT # at the first timestep return the decoder NLL, # otherwise KL(q(x_{t-1}|x_t,x_0) || p(x_{t-1}|x_t)) vb_losses = flow.where(times == 0, decoder_nll, kl) # weight the vb loss smaller, for stability, as in the paper (recommended 0.001) vb_loss = vb_losses.mean() * self.vb_loss_weight return loss + vb_loss @flow.no_grad() @eval_decorator def sample( self, image_embed=None, text=None, text_mask=None, text_encodings=None, batch_size=1, cond_scale=1.0, stop_at_unet_number=None, distributed=False, ): assert self.unconditional or exists( image_embed ), "image embed must be present on sampling from decoder unless if trained unconditionally" if not self.unconditional: batch_size = image_embed.shape[0] if exists(text) and not exists(text_encodings) and not self.unconditional: assert exists(self.clip) _, text_encodings, text_mask = self.clip.embed_text(text) assert not ( self.condition_on_text_encodings and not exists(text_encodings) ), "text or text encodings must be passed into decoder if specified" assert not ( not self.condition_on_text_encodings and exists(text_encodings) ), "decoder specified not to be conditioned on text, yet it is presented" img = None for ( unet_number, unet, vae, channel, image_size, predict_x_start, learned_variance, noise_scheduler, ) in tqdm( zip( range(1, len(self.unets) + 1), self.unets, self.vaes, self.sample_channels, self.image_sizes, self.predict_x_start, self.learned_variance, self.noise_schedulers, ) ): context = null_context() with context: lowres_cond_img = None shape = (batch_size, channel, image_size, image_size) if unet.lowres_cond: lowres_cond_img = self.to_lowres_cond(img, target_image_size=image_size) is_latent_diffusion = isinstance(vae, VQGanVAE) image_size = vae.get_encoded_fmap_size(image_size) shape = (batch_size, vae.encoded_dim, image_size, image_size) lowres_cond_img = maybe(vae.encode)(lowres_cond_img) img = self.p_sample_loop( unet, shape, image_embed=image_embed, text_encodings=text_encodings, text_mask=text_mask, cond_scale=cond_scale, predict_x_start=predict_x_start, learned_variance=learned_variance, clip_denoised=not is_latent_diffusion, lowres_cond_img=lowres_cond_img, is_latent_diffusion=is_latent_diffusion, noise_scheduler=noise_scheduler, ) img = vae.decode(img) if exists(stop_at_unet_number) and stop_at_unet_number == unet_number: break return img def forward( self, image, text=None, image_embed=None, text_encodings=None, text_mask=None, unet_number=None, return_lowres_cond_image=False, ): assert not ( len(self.unets) > 1 and not exists(unet_number) ), f"you must specify which unet you want trained, from a range of 1 to {len(self.unets)}," " if you are training cascading DDPM (multiple unets)" unet_number = default(unet_number, 1) unet_index = unet_number - 1 unet = self.get_unet(unet_number) vae = self.vaes[unet_index] noise_scheduler = self.noise_schedulers[unet_index] target_image_size = self.image_sizes[unet_index] predict_x_start = self.predict_x_start[unet_index] random_crop_size = self.random_crop_sizes[unet_index] learned_variance = self.learned_variance[unet_index] b, _, h, w, _, = ( *image.shape, image.device, ) check_shape(image, "b c h w", c=self.channels) assert h >= target_image_size and w >= target_image_size times = flow.randint( 0, noise_scheduler.num_timesteps, (b,), placement=get_default_placement(), sbp=get_default_sbp(), dtype=flow.long, ) if not exists(image_embed) and not self.unconditional: assert exists(self.clip), "if you want to derive CLIP image embeddings automatically, " "you must supply `clip` to the decoder on init" image_embed, _ = self.clip.embed_image(image) if exists(text) and not exists(text_encodings) and not self.unconditional: assert exists( self.clip ), "if you are passing in raw text, you need to supply `clip` to the decoder" _, text_encodings, text_mask = self.clip.embed_text(text) assert not ( self.condition_on_text_encodings and not exists(text_encodings) ), "text or text encodings must be passed into decoder if specified" assert not ( not self.condition_on_text_encodings and exists(text_encodings) ), "decoder specified not to be conditioned on text, yet it is presented" lowres_cond_img = ( self.to_lowres_cond( image, target_image_size=target_image_size, downsample_image_size=self.image_sizes[unet_index - 1], ) if unet_number > 1 else None ) image = resize_image_to(image, target_image_size) if exists(random_crop_size): aug = K.RandomCrop((random_crop_size, random_crop_size), p=1.0) # make sure low res conditioner and image both get augmented the same way image = aug(image) lowres_cond_img = aug(lowres_cond_img, params=aug._params) is_latent_diffusion = not isinstance(vae, NullVQGanVAE) vae.eval() with flow.no_grad(): image = vae.encode(image) lowres_cond_img = maybe(vae.encode)(lowres_cond_img) losses = self.p_losses( unet, image, times, image_embed=image_embed, text_encodings=text_encodings, text_mask=text_mask, lowres_cond_img=lowres_cond_img, predict_x_start=predict_x_start, learned_variance=learned_variance, is_latent_diffusion=is_latent_diffusion, noise_scheduler=noise_scheduler, ) if not return_lowres_cond_image: return losses return losses, lowres_cond_img # main class class DALLE2(nn.Module): def __init__(self, *, prior, decoder, prior_num_samples=2, **kwargs): super().__init__() # assert isinstance(prior, DiffusionPrior) # assert isinstance(decoder, Decoder) self.prior = prior self.decoder = decoder self.tokenizer = SimpleTokenizer() self.prior_num_samples = prior_num_samples self.decoder_need_text_cond = self.decoder.condition_on_text_encodings self.to_pil = T.ToPILImage() @flow.no_grad() @eval_decorator def forward(self, text, cond_scale=1.0, prior_cond_scale=1.0, return_pil_images=False): device = module_device(self) one_text = isinstance(text, str) or (not is_list_str(text) and text.shape[0] == 1) if isinstance(text, str) or is_list_str(text): text = [text] if not isinstance(text, (list, tuple)) else text text = self.tokenizer.tokenize(text).to(device) image_embed = self.prior.sample( text, num_samples_per_batch=self.prior_num_samples, cond_scale=prior_cond_scale ) text_cond = text if self.decoder_need_text_cond else None images = self.decoder.sample(image_embed, text=text_cond, cond_scale=cond_scale) if return_pil_images: images = list(map(self.to_pil, images.unbind(dim=0))) if one_text: return first(images) return images
83,430
31.821007
100
py
libai
libai-main/projects/DALLE2/dalle2/__init__.py
from .models import Unet, DALLE2, DiffusionPriorNetwork, DiffusionPrior, Decoder from ._clip import OpenAIClipAdapter, import_flow_clip
136
44.666667
80
py
libai
libai-main/projects/DALLE2/dalle2/vector_quantize_flow.py
# from https://github.com/lucidrains/vector_quantize_pytorch/vector_quantize_pytorch.py import oneflow as flow import oneflow.nn.functional as F from einops import rearrange, repeat from oneflow import einsum, nn from libai.utils import distributed def exists(val): return val is not None def default(val, d): return val if exists(val) else d def noop(*args, **kwargs): pass def l2norm(t): return F.normalize(t, p=2, dim=-1) def log(t, eps=1e-20): return flow.log(t.clamp(min=eps)) def uniform_init(*shape): t = flow.empty(shape) nn.init.kaiming_uniform_(t) return t def gumbel_noise(t): noise = flow.zeros_like(t).uniform_(0, 1) return -log(-log(noise)) def gumbel_sample(t, temperature=1.0, dim=-1): if temperature == 0: return t.argmax(dim=dim) return ((t / temperature) + gumbel_noise(t)).argmax(dim=dim) def ema_inplace(moving_avg, new, decay): moving_avg.data.mul_(decay).add_(new, alpha=(1 - decay)) def laplace_smoothing(x, n_categories, eps=1e-5): return (x + eps) / (x.sum() + n_categories * eps) def sample_vectors(samples, num): num_samples, device = samples.shape[0], samples.device if num_samples >= num: indices = flow.randperm(num_samples, device=device)[:num] else: indices = flow.randint(0, num_samples, (num,), device=device) return samples[indices] def batched_sample_vectors(samples, num): return flow.stack([sample_vectors(sample, num) for sample in samples.unbind(dim=0)], dim=0) def pad_shape(shape, size, dim=0): return [size if i == dim else s for i, s in enumerate(shape)] def sample_multinomial(total_count, probs): device = probs.device probs = probs.cpu() total_count = probs.new_full((), total_count) remainder = probs.new_ones(()) sample = flow.empty_like(probs, dtype=flow.long) for i, p in enumerate(probs): s = flow.binomial(total_count, p / remainder) sample[i] = s total_count -= s remainder -= p return sample.to(device) def all_gather_sizes(x, dim): size = flow.tensor(x.shape[dim], dtype=flow.long, device=x.device) all_sizes = [flow.empty_like(size) for _ in range(distributed.get_world_size())] distributed.all_gather(all_sizes, size) return flow.stack(all_sizes) def all_gather_variably_sized(x, sizes, dim=0): rank = distributed.get_rank() all_x = [] for i, size in enumerate(sizes): t = x if i == rank else x.new_empty(pad_shape(x.shape, size, dim)) distributed.broadcast(t, src=i, async_op=True) all_x.append(t) distributed.barrier() return all_x def sample_vectors_distributed(local_samples, num): rank = distributed.get_rank() all_num_samples = all_gather_sizes(local_samples, dim=0) if rank == 0: samples_per_rank = sample_multinomial(num, all_num_samples / all_num_samples.sum()) else: samples_per_rank = flow.empty_like(all_num_samples) distributed.broadcast(samples_per_rank, src=0) samples_per_rank = samples_per_rank.tolist() local_samples = batched_sample_vectors(local_samples, samples_per_rank[rank]) all_samples = all_gather_variably_sized(local_samples, samples_per_rank, dim=0) return flow.cat(all_samples, dim=0) def batched_bincount(x, *, minlength): batch, dtype, device = x.shape[0], x.dtype, x.device target = flow.zeros(batch, minlength, dtype=dtype, device=device) values = flow.ones_like(x) target.scatter_add_(-1, x, values) return target def kmeans( samples, num_clusters, num_iters=10, use_cosine_sim=False, sample_fn=batched_sample_vectors, all_reduce_fn=noop, ): num_codebooks, dim, dtype, _ = ( samples.shape[0], samples.shape[-1], samples.dtype, samples.device, ) means = sample_fn(samples, num_clusters) for _ in range(num_iters): if use_cosine_sim: dists = samples @ rearrange(means, "h n d -> h d n") else: dists = -flow.cdist(samples, means, p=2) buckets = flow.argmax(dists, dim=-1) bins = batched_bincount(buckets, minlength=num_clusters) all_reduce_fn(bins) zero_mask = bins == 0 bins_min_clamped = bins.masked_fill(zero_mask, 1) new_means = buckets.new_zeros(num_codebooks, num_clusters, dim, dtype=dtype) new_means.scatter_add_(1, repeat(buckets, "h n -> h n d", d=dim), samples) new_means = new_means / rearrange(bins_min_clamped, "... -> ... 1") all_reduce_fn(new_means) if use_cosine_sim: new_means = l2norm(new_means) means = flow.where(rearrange(zero_mask, "... -> ... 1"), means, new_means) return means, bins def batched_embedding(indices, embeds): batch, dim = indices.shape[1], embeds.shape[-1] indices = repeat(indices, "h b n -> h b n d", d=dim) embeds = repeat(embeds, "h c d -> h b c d", b=batch) return embeds.gather(2, indices) # regularization losses def orthgonal_loss_fn(t): # eq (2) from https://arxiv.org/abs/2112.00384 h, n = t.shape[:2] normed_codes = l2norm(t) identity = repeat(flow.eye(n, device=t.device), "i j -> h i j", h=h) cosine_sim = einsum("h i d, h j d -> h i j", normed_codes, normed_codes) return ((cosine_sim - identity) ** 2).sum() / (h * n ** 2) # distance types class EuclideanCodebook(nn.Module): def __init__( self, dim, codebook_size, num_codebooks=1, kmeans_init=False, kmeans_iters=10, decay=0.8, eps=1e-5, threshold_ema_dead_code=2, use_ddp=False, learnable_codebook=False, sample_codebook_temp=0, ): super().__init__() self.decay = decay init_fn = uniform_init if not kmeans_init else flow.zeros embed = init_fn(num_codebooks, codebook_size, dim) self.codebook_size = codebook_size self.num_codebooks = num_codebooks self.kmeans_iters = kmeans_iters self.eps = eps self.threshold_ema_dead_code = threshold_ema_dead_code self.sample_codebook_temp = sample_codebook_temp self.sample_fn = sample_vectors_distributed if use_ddp else batched_sample_vectors self.all_reduce_fn = distributed.all_reduce if use_ddp else noop self.register_buffer("initted", flow.Tensor([not kmeans_init])) self.register_buffer("cluster_size", flow.zeros(num_codebooks, codebook_size)) self.register_buffer("embed_avg", embed.clone()) self.learnable_codebook = learnable_codebook if learnable_codebook: self.embed = nn.Parameter(embed) else: self.register_buffer("embed", embed) def init_embed_(self, data): if self.initted: return embed, cluster_size = kmeans( data, self.codebook_size, self.kmeans_iters, sample_fn=self.sample_fn, all_reduce_fn=self.all_reduce_fn, ) self.embed.data.copy_(embed) self.embed_avg.data.copy_(embed.clone()) self.cluster_size.data.copy_(cluster_size) self.initted.data.copy_(flow.Tensor([True])) def replace(self, batch_samples, batch_mask): batch_samples = l2norm(batch_samples) for ind, (samples, mask) in enumerate( zip(batch_samples.unbind(dim=0), batch_mask.unbind(dim=0)) ): if not flow.any(mask): continue sampled = self.sample_fn(rearrange(samples, "... -> 1 ..."), mask.sum().item()) self.embed.data[ind][mask] = rearrange(sampled, "1 ... -> ...") def expire_codes_(self, batch_samples): if self.threshold_ema_dead_code == 0: return expired_codes = self.cluster_size < self.threshold_ema_dead_code if not flow.any(expired_codes): return batch_samples = rearrange(batch_samples, "h ... d -> h (...) d") self.replace(batch_samples, batch_mask=expired_codes) def forward(self, x): needs_codebook_dim = x.ndim < 4 x = x.float() if needs_codebook_dim: x = rearrange(x, "... -> 1 ...") shape, dtype = x.shape, x.dtype flatten = rearrange(x, "h ... d -> h (...) d") self.init_embed_(flatten) embed = self.embed if not self.learnable_codebook else self.embed.detach() dist = -flow.cdist(flatten, embed, p=2) embed_ind = gumbel_sample(dist, dim=-1, temperature=self.sample_codebook_temp) embed_onehot = F.one_hot(embed_ind, self.codebook_size).type(dtype) embed_ind = embed_ind.view(*shape[:-1]) quantize = batched_embedding(embed_ind, self.embed) if self.training: cluster_size = embed_onehot.sum(dim=1) self.all_reduce_fn(cluster_size) ema_inplace(self.cluster_size, cluster_size, self.decay) embed_sum = einsum("h n d, h n c -> h c d", flatten, embed_onehot) self.all_reduce_fn(embed_sum) cluster_size = ( laplace_smoothing(self.cluster_size, self.codebook_size, self.eps) * self.cluster_size.sum() ) embed_normalized = self.embed_avg / rearrange(cluster_size, "... -> ... 1") self.embed.data.copy_(embed_normalized) self.expire_codes_(x) if needs_codebook_dim: quantize, embed_ind = map(lambda t: rearrange(t, "1 ... -> ..."), (quantize, embed_ind)) return quantize, embed_ind class CosineSimCodebook(nn.Module): def __init__( self, dim, codebook_size, num_codebooks=1, kmeans_init=False, kmeans_iters=10, decay=0.8, eps=1e-5, threshold_ema_dead_code=2, use_ddp=False, learnable_codebook=False, sample_codebook_temp=0.0, ): super().__init__() self.decay = decay if not kmeans_init: embed = l2norm(uniform_init(num_codebooks, codebook_size, dim)) else: embed = flow.zeros(num_codebooks, codebook_size, dim) self.codebook_size = codebook_size self.num_codebooks = num_codebooks self.kmeans_iters = kmeans_iters self.eps = eps self.threshold_ema_dead_code = threshold_ema_dead_code self.sample_codebook_temp = sample_codebook_temp self.sample_fn = sample_vectors_distributed if use_ddp else batched_sample_vectors self.all_reduce_fn = distributed.all_reduce if use_ddp else noop self.register_buffer("initted", flow.Tensor([not kmeans_init])) self.register_buffer("cluster_size", flow.zeros(num_codebooks, codebook_size)) self.learnable_codebook = learnable_codebook if learnable_codebook: self.embed = nn.Parameter(embed) else: self.register_buffer("embed", embed) def init_embed_(self, data): if self.initted: return embed, cluster_size = kmeans( data, self.codebook_size, self.kmeans_iters, use_cosine_sim=True, sample_fn=self.sample_fn, all_reduce_fn=self.all_reduce_fn, ) self.embed.data.copy_(embed) self.cluster_size.data.copy_(cluster_size) self.initted.data.copy_(flow.Tensor([True])) def replace(self, batch_samples, batch_mask): batch_samples = l2norm(batch_samples) for ind, (samples, mask) in enumerate( zip(batch_samples.unbind(dim=0), batch_mask.unbind(dim=0)) ): if not flow.any(mask): continue sampled = self.sample_fn(rearrange(samples, "... -> 1 ..."), mask.sum().item()) self.embed.data[ind][mask] = rearrange(sampled, "1 ... -> ...") def expire_codes_(self, batch_samples): if self.threshold_ema_dead_code == 0: return expired_codes = self.cluster_size < self.threshold_ema_dead_code if not flow.any(expired_codes): return batch_samples = rearrange(batch_samples, "h ... d -> h (...) d") self.replace(batch_samples, batch_mask=expired_codes) def forward(self, x): needs_codebook_dim = x.ndim < 4 x = x.float() if needs_codebook_dim: x = rearrange(x, "... -> 1 ...") shape, dtype = x.shape, x.dtype flatten = rearrange(x, "h ... d -> h (...) d") flatten = l2norm(flatten) self.init_embed_(flatten) embed = self.embed if not self.learnable_codebook else self.embed.detach() embed = l2norm(embed) dist = einsum("h n d, h c d -> h n c", flatten, embed) embed_ind = gumbel_sample(dist, dim=-1, temperature=self.sample_codebook_temp) embed_onehot = F.one_hot(embed_ind, self.codebook_size).type(dtype) embed_ind = embed_ind.view(*shape[:-1]) quantize = batched_embedding(embed_ind, self.embed) if self.training: bins = embed_onehot.sum(dim=1) self.all_reduce_fn(bins) ema_inplace(self.cluster_size, bins, self.decay) zero_mask = bins == 0 bins = bins.masked_fill(zero_mask, 1.0) embed_sum = einsum("h n d, h n c -> h c d", flatten, embed_onehot) self.all_reduce_fn(embed_sum) embed_normalized = embed_sum / rearrange(bins, "... -> ... 1") embed_normalized = l2norm(embed_normalized) embed_normalized = flow.where( rearrange(zero_mask, "... -> ... 1"), embed, embed_normalized ) ema_inplace(self.embed, embed_normalized, self.decay) self.expire_codes_(x) if needs_codebook_dim: quantize, embed_ind = map(lambda t: rearrange(t, "1 ... -> ..."), (quantize, embed_ind)) return quantize, embed_ind # main class class VectorQuantize(nn.Module): def __init__( self, dim, codebook_size, codebook_dim=None, heads=1, separate_codebook_per_head=False, decay=0.8, eps=1e-5, kmeans_init=False, kmeans_iters=10, use_cosine_sim=False, threshold_ema_dead_code=0, channel_last=True, accept_image_fmap=False, commitment_weight=1.0, orthogonal_reg_weight=0.0, orthogonal_reg_active_codes_only=False, orthogonal_reg_max_codes=None, sample_codebook_temp=0.0, sync_codebook=False, ): super().__init__() self.heads = heads self.separate_codebook_per_head = separate_codebook_per_head codebook_dim = default(codebook_dim, dim) codebook_input_dim = codebook_dim * heads requires_projection = codebook_input_dim != dim self.project_in = ( nn.Linear(dim, codebook_input_dim) if requires_projection else nn.Identity() ) self.project_out = ( nn.Linear(codebook_input_dim, dim) if requires_projection else nn.Identity() ) self.eps = eps self.commitment_weight = commitment_weight has_codebook_orthogonal_loss = orthogonal_reg_weight > 0 self.orthogonal_reg_weight = orthogonal_reg_weight self.orthogonal_reg_active_codes_only = orthogonal_reg_active_codes_only self.orthogonal_reg_max_codes = orthogonal_reg_max_codes codebook_class = EuclideanCodebook if not use_cosine_sim else CosineSimCodebook self._codebook = codebook_class( dim=codebook_dim, num_codebooks=heads if separate_codebook_per_head else 1, codebook_size=codebook_size, kmeans_init=kmeans_init, kmeans_iters=kmeans_iters, decay=decay, eps=eps, threshold_ema_dead_code=threshold_ema_dead_code, use_ddp=sync_codebook, learnable_codebook=has_codebook_orthogonal_loss, sample_codebook_temp=sample_codebook_temp, ) self.codebook_size = codebook_size self.accept_image_fmap = accept_image_fmap self.channel_last = channel_last @property def codebook(self): return self._codebook.embed def forward(self, x): _, device, heads, is_multiheaded, _ = ( x.shape, x.device, self.heads, self.heads > 1, self.codebook_size, ) need_transpose = not self.channel_last and not self.accept_image_fmap if self.accept_image_fmap: height, width = x.shape[-2:] x = rearrange(x, "b c h w -> b (h w) c") if need_transpose: x = rearrange(x, "b d n -> b n d") x = self.project_in(x) if is_multiheaded: ein_rhs_eq = "h b n d" if self.separate_codebook_per_head else "1 (b h) n d" x = rearrange(x, f"b n (h d) -> {ein_rhs_eq}", h=heads) quantize, embed_ind = self._codebook(x) if self.training: quantize = x + (quantize - x).detach() loss = flow.tensor([0.0], device=device, requires_grad=self.training) if self.training: if self.commitment_weight > 0: commit_loss = F.mse_loss(quantize.detach(), x) loss = loss + commit_loss * self.commitment_weight if self.orthogonal_reg_weight > 0: codebook = self.codebook if self.orthogonal_reg_active_codes_only: # only calculate orthogonal loss for the activated codes for this batch unique_code_ids = flow.unique(embed_ind) codebook = codebook[unique_code_ids] num_codes = codebook.shape[0] if ( exists(self.orthogonal_reg_max_codes) and num_codes > self.orthogonal_reg_max_codes ): rand_ids = flow.randperm(num_codes, device=device)[ : self.orthogonal_reg_max_codes ] codebook = codebook[rand_ids] orthogonal_reg_loss = orthgonal_loss_fn(codebook) loss = loss + orthogonal_reg_loss * self.orthogonal_reg_weight if is_multiheaded: if self.separate_codebook_per_head: quantize = rearrange(quantize, "h b n d -> b n (h d)", h=heads) embed_ind = rearrange(embed_ind, "h b n -> b n h", h=heads) else: quantize = rearrange(quantize, "1 (b h) n d -> b n (h d)", h=heads) embed_ind = rearrange(embed_ind, "1 (b h) n -> b n h", h=heads) quantize = self.project_out(quantize) if need_transpose: quantize = rearrange(quantize, "b n d -> b d n") if self.accept_image_fmap: quantize = rearrange(quantize, "b (h w) c -> b c h w", h=height, w=width) embed_ind = rearrange(embed_ind, "b (h w) ... -> b h w ...", h=height, w=width) return quantize, embed_ind, loss
19,209
30.033926
100
py
libai
libai-main/projects/DALLE2/dalle2/einops_exts.py
# From https://github.com/arogozhnikov/einops/blob/master/einops/layers/oneflow.py import re from functools import wraps import oneflow as flow from einops import rearrange, reduce, repeat from einops._backends import AbstractBackend from einops.layers import RearrangeMixin from oneflow import nn class Rearrange(RearrangeMixin, flow.nn.Module): def forward(self, input): return self._apply_recipe(input) class OneFlowBackend(AbstractBackend): framework_name = "oneflow" def __init__(self): import oneflow as flow self.flow = flow def is_appropriate_type(self, tensor): return isinstance(tensor, self.flow.Tensor) def from_numpy(self, x): variable = self.flow.from_numpy(x) if self.is_float_type(variable): # attach grad only to floating types variable.requires_grad = True return variable def to_numpy(self, x): return x.detach().cpu().numpy() def arange(self, start, stop): return self.flow.arange(start, stop, dtype=self.flow.int64) def reduce(self, x, operation, reduced_axes): for axis in sorted(reduced_axes, reverse=True): if operation == "min": x, _ = x.min(dim=axis) elif operation == "max": x, _ = x.max(dim=axis) elif operation in ["sum", "mean", "prod"]: x = getattr(x, operation)(dim=axis) else: raise NotImplementedError("Unknown reduction ", operation) return x def transpose(self, x, axes): return x.permute(axes) def stack_on_zeroth_dimension(self, tensors: list): return self.flow.stack(tensors) def add_axes(self, x, n_axes, pos2len): repeats = [-1] * n_axes for axis_position, axis_length in pos2len.items(): x = self.add_axis(x, axis_position) repeats[axis_position] = axis_length return x.expand(*repeats) def tile(self, x, repeats): return x.repeat(repeats) def add_axis(self, x, new_position): return self.flow.unsqueeze(x, new_position) def is_float_type(self, x): return x.dtype in [self.flow.float16, self.flow.float32, self.flow.float64] def einsum(self, pattern, *x): return self.flow.einsum(pattern, *x) # From https://github.com/lucidrains/einops-exts/tree/main/einops_exts class EinopsToAndFrom(nn.Module): def __init__(self, from_einops, to_einops, fn): super().__init__() self.from_einops = from_einops self.to_einops = to_einops self.fn = fn def forward(self, x, **kwargs): shape = x.shape reconstitute_kwargs = dict(tuple(zip(self.from_einops.split(" "), shape))) x = rearrange(x, f"{self.from_einops} -> {self.to_einops}") x = self.fn(x, **kwargs) x = rearrange(x, f"{self.to_einops} -> {self.from_einops}", **reconstitute_kwargs) return x # checking shape # @nils-werner # https://github.com/arogozhnikov/einops/issues/168#issuecomment-1042933838 def check_shape(tensor, pattern, **kwargs): return rearrange(tensor, f"{pattern} -> {pattern}", **kwargs) # do same einops operations on a list of tensors def _many(fn): @wraps(fn) def inner(tensors, pattern, **kwargs): return (fn(tensor, pattern, **kwargs) for tensor in tensors) return inner # do einops with unflattening of anonymously named dimensions # (...flattened) -> ...flattened def _with_anon_dims(fn): @wraps(fn) def inner(tensor, pattern, **kwargs): regex = r"(\.\.\.[a-zA-Z]+)" matches = re.findall(regex, pattern) def get_anon_dim_name(t): return t.lstrip("...") dim_prefixes = tuple(map(get_anon_dim_name, set(matches))) update_kwargs_dict = dict() for prefix in dim_prefixes: assert prefix in kwargs, f'dimension list "{prefix}" was not passed in' dim_list = kwargs[prefix] assert isinstance( dim_list, (list, tuple) ), f'dimension list "{prefix}" needs to be a tuple of list of dimensions' dim_names = list(map(lambda ind: f"{prefix}{ind}", range(len(dim_list)))) update_kwargs_dict[prefix] = dict(zip(dim_names, dim_list)) def sub_with_anonymous_dims(t): dim_name_prefix = get_anon_dim_name(t.groups()[0]) return " ".join(update_kwargs_dict[dim_name_prefix].keys()) pattern_new = re.sub(regex, sub_with_anonymous_dims, pattern) for prefix, update_dict in update_kwargs_dict.items(): del kwargs[prefix] kwargs.update(update_dict) return fn(tensor, pattern_new, **kwargs) return inner # generate all helper functions rearrange_many = _many(rearrange) repeat_many = _many(repeat) reduce_many = _many(reduce) rearrange_with_anon_dims = _with_anon_dims(rearrange) repeat_with_anon_dims = _with_anon_dims(repeat) reduce_with_anon_dims = _with_anon_dims(reduce)
5,044
28.852071
90
py
libai
libai-main/projects/DALLE2/dalle2/model_weights/download_utils.py
import logging import os from libai.utils.file_utils import download_file logger = logging.getLogger(__name__) # pylint: disable=invalid-name url_map = { "prior": "https://huggingface.co/nousr/conditioned-prior/resolve/main/vit-l-14/prior_aes_finetune.pth", # noqa "decoder": "https://huggingface.co/laion/DALLE2-PyTorch/resolve/main/decoder/1.5B_laion2B/latest.pth", # noqa "clip": "https://openaipublic.azureedge.net/clip/models/b8cca3fd41ae0c99ba7e8951adf17d267cdb84cd88be6f7c2e0eca1737a03836/ViT-L-14.pt", # noqa "bpe_vocab": "https://oneflow-static.oss-cn-beijing.aliyuncs.com/libai/clip/bpe_simple_vocab_16e6.txt.gz", # noqa "swinir": "https://github.com/JingyunLiang/SwinIR/releases/download/v0.0/", } def _download_if_not_exist(path, name): if os.path.exists(path): logger.info((f"using {name}'s weight at {path}")) return if name == "swinir": os.makedirs(os.path.dirname(path), exist_ok=True) download_file(path, url_map[name] + os.path.basename(path)) return os.makedirs(os.path.dirname(path), exist_ok=True) download_file(path, url_map[name]) def download_dalle2_weights(cfg): if not os.path.exists("./dalle2/data/bpe_simple_vocab_16e6.txt.gz"): os.makedirs("./dalle2/data", exist_ok=True) download_file("./dalle2/data/bpe_simple_vocab_16e6.txt.gz", url_map["bpe_vocab"]) _download_if_not_exist(cfg.swinir.swinir_path, "swinir") _download_if_not_exist(cfg.model.prior_weight_path, "prior") _download_if_not_exist(cfg.model.decoder_weight_path, "decoder") _download_if_not_exist(cfg.model.prior.clip.name, "clip")
1,649
42.421053
146
py
libai
libai-main/projects/DALLE2/configs/dalle2_config.py
from libai.config import LazyCall from configs.common.train import train from dalle2.models import DiffusionPrior, DiffusionPriorNetwork, Unet, Decoder, DALLE2 from dalle2._clip import OpenAIClipAdapter from omegaconf import DictConfig clip = LazyCall(OpenAIClipAdapter)(name="") swinir = DictConfig({"swinir_path": None}) prior = LazyCall(DiffusionPrior)( net=LazyCall(DiffusionPriorNetwork)( dim=768, depth=24, num_timesteps=1000, max_text_len=77, num_time_embeds=1, num_image_embeds=1, num_text_embeds=1, dim_head=64, heads=32, ff_mult=4, attn_dropout=0.05, ff_dropout=0.05, normformer=True, ), clip=clip, image_embed_dim=768, timesteps=1000, cond_drop_prob=0.1, loss_type="l2", condition_on_text_encodings=True, ) unet1 = LazyCall(Unet)( dim=320, image_embed_dim=768, text_embed_dim=768, cond_dim=512, channels=3, dim_mults=(1, 2, 3, 4), num_resnet_blocks=4, attn_heads=8, attn_dim_head=64, sparse_attn=True, memory_efficient=True, cond_on_text_encodings=True, self_attn=[False, True, True, True], ) decoder = LazyCall(Decoder)( unet=(unet1,), image_sizes=[ 64, ], clip=None, channels=3, timesteps=1000, loss_type="l2", beta_schedule=["cosine"], learned_variance=True, ) model = LazyCall(DALLE2)( prior=prior, decoder=decoder, prior_weight_path="", decoder_weight_path="", )
1,532
21.217391
86
py
libai
libai-main/projects/DALLE2/swinir/utils.py
# ----------------------------------------------------------------------------------- # from # https://github.com/rwightman/pytorch-image-models/blob/master/timm/models/layers/weight_init.py # ----------------------------------------------------------------------------------- import collections.abc import math import warnings from itertools import repeat import oneflow as flow import oneflow.nn as nn def _no_grad_trunc_normal_(tensor, mean, std, a, b): # Cut & paste from Pytorch official master until it's in a few official releases - RW # Method based on https://people.sc.fsu.edu/~jburkardt/presentations/truncated_normal.pdf def norm_cdf(x): # Computes standard normal cumulative distribution function return (1.0 + math.erf(x / math.sqrt(2.0))) / 2.0 if (mean < a - 2 * std) or (mean > b + 2 * std): warnings.warn( "mean is more than 2 std from [a, b] in nn.init.trunc_normal_. " "The distribution of values may be incorrect.", stacklevel=2, ) with flow.no_grad(): # Values are generated by using a truncated uniform distribution and # then using the inverse CDF for the normal distribution. # Get upper and lower cdf values l = norm_cdf((a - mean) / std) u = norm_cdf((b - mean) / std) # Uniformly fill tensor with values from [l, u], then translate to # [2l-1, 2u-1]. tensor.uniform_(2 * l - 1, 2 * u - 1) # Use inverse cdf transform for normal distribution to get truncated # standard normal tensor.erfinv_() # Transform to proper mean, std tensor.mul_(std * math.sqrt(2.0)) tensor.add_(mean) # Clamp to ensure it's in the proper range tensor.clamp_(min=a, max=b) return tensor def trunc_normal_(tensor, mean=0.0, std=1.0, a=-2.0, b=2.0): # type: (flow.Tensor, float, float, float, float) -> flow.Tensor r"""Fills the input Tensor with values drawn from a truncated normal distribution. The values are effectively drawn from the normal distribution :math:`\mathcal{N}(\text{mean}, \text{std}^2)` with values outside :math:`[a, b]` redrawn until they are within the bounds. The method used for generating the random values works best when :math:`a \leq \text{mean} \leq b`. NOTE: this impl is similar to the Pytorch trunc_normal_, the bounds [a, b] are applied while sampling the normal with mean/std applied, therefore a, b args should be adjusted to match the range of mean, std args. Args: tensor: an n-dimensional `torch.Tensor` mean: the mean of the normal distribution std: the standard deviation of the normal distribution a: the minimum cutoff value b: the maximum cutoff value Examples: >>> w = torch.empty(3, 5) >>> nn.init.trunc_normal_(w) """ return _no_grad_trunc_normal_(tensor, mean, std, a, b) # ----------------------------------------------------------------------------------- # from https://github.com/rwightman/pytorch-image-models/blob/master/timm/models/layers/helpers.py # ----------------------------------------------------------------------------------- # From Pytorch internals def _ntuple(n): def parse(x): if isinstance(x, collections.abc.Iterable) and not isinstance(x, str): return x return tuple(repeat(x, n)) return parse to_1tuple = _ntuple(1) to_2tuple = _ntuple(2) to_3tuple = _ntuple(3) to_4tuple = _ntuple(4) to_ntuple = _ntuple # ----------------------------------------------------------------------------------- # from https://github.com/rwightman/pytorch-image-models/blob/master/timm/models/layers/drop.py # ----------------------------------------------------------------------------------- def drop_path(x, drop_prob: float = 0.0, training: bool = False, scale_by_keep: bool = True): """Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). This is the same as the DropConnect impl I created for EfficientNet, etc networks, however, the original name is misleading as 'Drop Connect' is a different form of dropout in a separate paper... See discussion: https://github.com/tensortorch/tpu/issues/494#issuecomment-532968956 ... I've opted for changing the layer and argument names to 'drop path' rather than mix DropConnect as a layer name and use 'survival rate' as the argument. """ if drop_prob == 0.0 or not training: return x keep_prob = 1 - drop_prob shape = (x.shape[0],) + (1,) * (x.ndim - 1) # work with diff dim tensors, not just 2D ConvNets random_tensor = x.new_empty(shape).bernoulli_(keep_prob) if keep_prob > 0.0 and scale_by_keep: random_tensor.div_(keep_prob) return x * random_tensor class DropPath(nn.Module): """Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).""" def __init__(self, drop_prob: float = 0.0, scale_by_keep: bool = True): super(DropPath, self).__init__() self.drop_prob = drop_prob self.scale_by_keep = scale_by_keep def forward(self, x): return drop_path(x, self.drop_prob, self.training, self.scale_by_keep) def extra_repr(self): return f"drop_prob={round(self.drop_prob,3):0.3f}"
5,374
39.413534
99
py
libai
libai-main/projects/DALLE2/swinir/models.py
# ----------------------------------------------------------------------------------- # SwinIR: Image Restoration Using Swin Transformer, https://arxiv.org/abs/2108.10257 # Originally Written by Ze Liu, Modified by Jingyun Liang. # ----------------------------------------------------------------------------------- # code from https://github.com/JingyunLiang/SwinIR/blob/main/models/network_swinir.py import math import oneflow as flow import oneflow.nn as nn import oneflow.nn.functional as F from oneflow.utils import checkpoint from .utils import DropPath, to_2tuple, trunc_normal_ class Mlp(nn.Module): def __init__( self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.0 ): super().__init__() out_features = out_features or in_features hidden_features = hidden_features or in_features self.fc1 = nn.Linear(in_features, hidden_features) self.act = act_layer() self.fc2 = nn.Linear(hidden_features, out_features) self.drop = nn.Dropout(drop) def forward(self, x): x = self.fc1(x) x = self.act(x) x = self.drop(x) x = self.fc2(x) x = self.drop(x) return x def window_partition(x, window_size): """ Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C) """ B, H, W, C = x.shape x = x.view(B, H // window_size, window_size, W // window_size, window_size, C) windows = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, window_size, window_size, C) return windows def window_reverse(windows, window_size, H, W): """ Args: windows: (num_windows*B, window_size, window_size, C) window_size (int): Window size H (int): Height of image W (int): Width of image Returns: x: (B, H, W, C) """ B = int(windows.shape[0] / (H * W / window_size / window_size)) x = windows.view(B, H // window_size, W // window_size, window_size, window_size, -1) x = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(B, H, W, -1) return x class WindowAttention(nn.Module): r"""Window based multi-head self attention (W-MSA) module with relative position bias. It supports both of shifted and non-shifted window. Args: dim (int): Number of input channels. window_size (tuple[int]): The height and width of the window. num_heads (int): Number of attention heads. qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set attn_drop (float, optional): Dropout ratio of attention weight. Default: 0.0 proj_drop (float, optional): Dropout ratio of output. Default: 0.0 """ def __init__( self, dim, window_size, num_heads, qkv_bias=True, qk_scale=None, attn_drop=0.0, proj_drop=0.0, ): super().__init__() self.dim = dim self.window_size = window_size # Wh, Ww self.num_heads = num_heads head_dim = dim // num_heads self.scale = qk_scale or head_dim ** -0.5 # define a parameter table of relative position bias self.relative_position_bias_table = nn.Parameter( flow.zeros((2 * window_size[0] - 1) * (2 * window_size[1] - 1), num_heads) ) # 2*Wh-1 * 2*Ww-1, nH # get pair-wise relative position index for each token inside the window coords_h = flow.arange(self.window_size[0]) coords_w = flow.arange(self.window_size[1]) coords = flow.stack(flow.meshgrid([coords_h, coords_w])) # 2, Wh, Ww coords_flatten = flow.flatten(coords, 1) # 2, Wh*Ww relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :] # 2, Wh*Ww, Wh*Ww relative_coords = relative_coords.permute(1, 2, 0).contiguous() # Wh*Ww, Wh*Ww, 2 relative_coords[:, :, 0] += self.window_size[0] - 1 # shift to start from 0 relative_coords[:, :, 1] += self.window_size[1] - 1 relative_coords[:, :, 0] *= 2 * self.window_size[1] - 1 relative_position_index = relative_coords.sum(-1) # Wh*Ww, Wh*Ww self.register_buffer("relative_position_index", relative_position_index) self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias) self.attn_drop = nn.Dropout(attn_drop) self.proj = nn.Linear(dim, dim) self.proj_drop = nn.Dropout(proj_drop) trunc_normal_(self.relative_position_bias_table, std=0.02) self.softmax = nn.Softmax(dim=-1) def forward(self, x, mask=None): """ Args: x: input features with shape of (num_windows*B, N, C) mask: (0/-inf) mask with shape of (num_windows, Wh*Ww, Wh*Ww) or None """ B_, N, C = x.shape qkv = ( self.qkv(x) .reshape(B_, N, 3, self.num_heads, C // self.num_heads) .permute(2, 0, 3, 1, 4) ) q, k, v = qkv[0], qkv[1], qkv[2] # make torchscript happy (cannot use tensor as tuple) q = q * self.scale attn = q @ k.transpose(-2, -1) relative_position_bias = self.relative_position_bias_table[ self.relative_position_index.view(-1) ].view( self.window_size[0] * self.window_size[1], self.window_size[0] * self.window_size[1], -1 ) # Wh*Ww,Wh*Ww,nH relative_position_bias = relative_position_bias.permute( 2, 0, 1 ).contiguous() # nH, Wh*Ww, Wh*Ww attn = attn + relative_position_bias.unsqueeze(0) if mask is not None: nW = mask.shape[0] attn = attn.view(B_ // nW, nW, self.num_heads, N, N) + mask.unsqueeze(1).unsqueeze(0) attn = attn.view(-1, self.num_heads, N, N) attn = self.softmax(attn) else: attn = self.softmax(attn) attn = self.attn_drop(attn) x = (attn @ v).transpose(1, 2).reshape(B_, N, C) x = self.proj(x) x = self.proj_drop(x) return x def extra_repr(self) -> str: return f"dim={self.dim}, window_size={self.window_size}, num_heads={self.num_heads}" def flops(self, N): # calculate flops for 1 window with token length of N flops = 0 # qkv = self.qkv(x) flops += N * self.dim * 3 * self.dim # attn = (q @ k.transpose(-2, -1)) flops += self.num_heads * N * (self.dim // self.num_heads) * N # x = (attn @ v) flops += self.num_heads * N * N * (self.dim // self.num_heads) # x = self.proj(x) flops += N * self.dim * self.dim return flops class SwinTransformerBlock(nn.Module): r"""Swin Transformer Block. Args: dim (int): Number of input channels. input_resolution (tuple[int]): Input resulotion. num_heads (int): Number of attention heads. window_size (int): Window size. shift_size (int): Shift size for SW-MSA. mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set. drop (float, optional): Dropout rate. Default: 0.0 attn_drop (float, optional): Attention dropout rate. Default: 0.0 drop_path (float, optional): Stochastic depth rate. Default: 0.0 act_layer (nn.Module, optional): Activation layer. Default: nn.GELU norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm """ def __init__( self, dim, input_resolution, num_heads, window_size=7, shift_size=0, mlp_ratio=4.0, qkv_bias=True, qk_scale=None, drop=0.0, attn_drop=0.0, drop_path=0.0, act_layer=nn.GELU, norm_layer=nn.LayerNorm, ): super().__init__() self.dim = dim self.input_resolution = input_resolution self.num_heads = num_heads self.window_size = window_size self.shift_size = shift_size self.mlp_ratio = mlp_ratio if min(self.input_resolution) <= self.window_size: # if window size is larger than input resolution, we don't partition windows self.shift_size = 0 self.window_size = min(self.input_resolution) assert 0 <= self.shift_size < self.window_size, "shift_size must in 0-window_size" self.norm1 = norm_layer(dim) self.attn = WindowAttention( dim, window_size=to_2tuple(self.window_size), num_heads=num_heads, qkv_bias=qkv_bias, qk_scale=qk_scale, attn_drop=attn_drop, proj_drop=drop, ) self.drop_path = DropPath(drop_path) if drop_path > 0.0 else nn.Identity() self.norm2 = norm_layer(dim) mlp_hidden_dim = int(dim * mlp_ratio) self.mlp = Mlp( in_features=dim, hidden_features=mlp_hidden_dim, act_layer=act_layer, drop=drop ) if self.shift_size > 0: attn_mask = self.calculate_mask(self.input_resolution) else: attn_mask = None self.register_buffer("attn_mask", attn_mask) def calculate_mask(self, x_size): # calculate attention mask for SW-MSA H, W = x_size img_mask = flow.zeros((1, H, W, 1)) # 1 H W 1 h_slices = ( slice(0, -self.window_size), slice(-self.window_size, -self.shift_size), slice(-self.shift_size, None), ) w_slices = ( slice(0, -self.window_size), slice(-self.window_size, -self.shift_size), slice(-self.shift_size, None), ) cnt = 0 for h in h_slices: for w in w_slices: img_mask[:, h, w, :] = cnt cnt += 1 mask_windows = window_partition( img_mask, self.window_size ) # nW, window_size, window_size, 1 mask_windows = mask_windows.view(-1, self.window_size * self.window_size) attn_mask = mask_windows.unsqueeze(1) - mask_windows.unsqueeze(2) attn_mask = attn_mask.masked_fill(attn_mask != 0, float(-100.0)).masked_fill( attn_mask == 0, float(0.0) ) return attn_mask def forward(self, x, x_size): H, W = x_size B, L, C = x.shape # assert L == H * W, "input feature has wrong size" shortcut = x x = self.norm1(x) x = x.view(B, H, W, C) # cyclic shift if self.shift_size > 0: shifted_x = flow.roll(x, shifts=(-self.shift_size, -self.shift_size), dims=(1, 2)) else: shifted_x = x # partition windows x_windows = window_partition( shifted_x, self.window_size ) # nW*B, window_size, window_size, C x_windows = x_windows.view( -1, self.window_size * self.window_size, C ) # nW*B, window_size*window_size, C # W-MSA/SW-MSA (to be compatible for testing on images # whose shapes are the multiple of window size if self.input_resolution == x_size: attn_windows = self.attn( x_windows, mask=self.attn_mask ) # nW*B, window_size*window_size, C else: attn_windows = self.attn(x_windows, mask=self.calculate_mask(x_size).to(x.device)) # merge windows attn_windows = attn_windows.view(-1, self.window_size, self.window_size, C) shifted_x = window_reverse(attn_windows, self.window_size, H, W) # B H' W' C # reverse cyclic shift if self.shift_size > 0: x = flow.roll(shifted_x, shifts=(self.shift_size, self.shift_size), dims=(1, 2)) else: x = shifted_x x = x.view(B, H * W, C) # FFN x = shortcut + self.drop_path(x) x = x + self.drop_path(self.mlp(self.norm2(x))) return x def extra_repr(self) -> str: return ( f"dim={self.dim}, input_resolution={self.input_resolution}, " f"num_heads={self.num_heads}, window_size={self.window_size}," f"shift_size={self.shift_size}, mlp_ratio={self.mlp_ratio}" ) def flops(self): flops = 0 H, W = self.input_resolution # norm1 flops += self.dim * H * W # W-MSA/SW-MSA nW = H * W / self.window_size / self.window_size flops += nW * self.attn.flops(self.window_size * self.window_size) # mlp flops += 2 * H * W * self.dim * self.dim * self.mlp_ratio # norm2 flops += self.dim * H * W return flops class PatchMerging(nn.Module): r"""Patch Merging Layer. Args: input_resolution (tuple[int]): Resolution of input feature. dim (int): Number of input channels. norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm """ def __init__(self, input_resolution, dim, norm_layer=nn.LayerNorm): super().__init__() self.input_resolution = input_resolution self.dim = dim self.reduction = nn.Linear(4 * dim, 2 * dim, bias=False) self.norm = norm_layer(4 * dim) def forward(self, x): """ x: B, H*W, C """ H, W = self.input_resolution B, L, C = x.shape assert L == H * W, "input feature has wrong size" assert H % 2 == 0 and W % 2 == 0, f"x size ({H}*{W}) are not even." x = x.view(B, H, W, C) x0 = x[:, 0::2, 0::2, :] # B H/2 W/2 C x1 = x[:, 1::2, 0::2, :] # B H/2 W/2 C x2 = x[:, 0::2, 1::2, :] # B H/2 W/2 C x3 = x[:, 1::2, 1::2, :] # B H/2 W/2 C x = flow.cat([x0, x1, x2, x3], -1) # B H/2 W/2 4*C x = x.view(B, -1, 4 * C) # B H/2*W/2 4*C x = self.norm(x) x = self.reduction(x) return x def extra_repr(self) -> str: return f"input_resolution={self.input_resolution}, dim={self.dim}" def flops(self): H, W = self.input_resolution flops = H * W * self.dim flops += (H // 2) * (W // 2) * 4 * self.dim * 2 * self.dim return flops class BasicLayer(nn.Module): """A basic Swin Transformer layer for one stage. Args: dim (int): Number of input channels. input_resolution (tuple[int]): Input resolution. depth (int): Number of blocks. num_heads (int): Number of attention heads. window_size (int): Local window size. mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set. drop (float, optional): Dropout rate. Default: 0.0 attn_drop (float, optional): Attention dropout rate. Default: 0.0 drop_path (float | tuple[float], optional): Stochastic depth rate. Default: 0.0 norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm downsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default: None use_checkpoint (bool): Whether to use checkpointing to save memory. Default: False. """ def __init__( self, dim, input_resolution, depth, num_heads, window_size, mlp_ratio=4.0, qkv_bias=True, qk_scale=None, drop=0.0, attn_drop=0.0, drop_path=0.0, norm_layer=nn.LayerNorm, downsample=None, use_checkpoint=False, ): super().__init__() self.dim = dim self.input_resolution = input_resolution self.depth = depth self.use_checkpoint = use_checkpoint # build blocks self.blocks = nn.ModuleList( [ SwinTransformerBlock( dim=dim, input_resolution=input_resolution, num_heads=num_heads, window_size=window_size, shift_size=0 if (i % 2 == 0) else window_size // 2, mlp_ratio=mlp_ratio, qkv_bias=qkv_bias, qk_scale=qk_scale, drop=drop, attn_drop=attn_drop, drop_path=drop_path[i] if isinstance(drop_path, list) else drop_path, norm_layer=norm_layer, ) for i in range(depth) ] ) # patch merging layer if downsample is not None: self.downsample = downsample(input_resolution, dim=dim, norm_layer=norm_layer) else: self.downsample = None def forward(self, x, x_size): for blk in self.blocks: if self.use_checkpoint: x = checkpoint.checkpoint(blk, x, x_size) else: x = blk(x, x_size) if self.downsample is not None: x = self.downsample(x) return x def extra_repr(self) -> str: return f"dim={self.dim}, input_resolution={self.input_resolution}, depth={self.depth}" def flops(self): flops = 0 for blk in self.blocks: flops += blk.flops() if self.downsample is not None: flops += self.downsample.flops() return flops class RSTB(nn.Module): """Residual Swin Transformer Block (RSTB). Args: dim (int): Number of input channels. input_resolution (tuple[int]): Input resolution. depth (int): Number of blocks. num_heads (int): Number of attention heads. window_size (int): Local window size. mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. qkv_bias (bool, optional): If True, add a learnable bias to query, key, value. Default: True qk_scale (float | None, optional): Override default qk scale of head_dim ** -0.5 if set. drop (float, optional): Dropout rate. Default: 0.0 attn_drop (float, optional): Attention dropout rate. Default: 0.0 drop_path (float | tuple[float], optional): Stochastic depth rate. Default: 0.0 norm_layer (nn.Module, optional): Normalization layer. Default: nn.LayerNorm downsample (nn.Module | None, optional): Downsample layer at the end of the layer. Default: None use_checkpoint (bool): Whether to use checkpointing to save memory. Default: False. img_size: Input image size. patch_size: Patch size. resi_connection: The convolutional block before residual connection. """ def __init__( self, dim, input_resolution, depth, num_heads, window_size, mlp_ratio=4.0, qkv_bias=True, qk_scale=None, drop=0.0, attn_drop=0.0, drop_path=0.0, norm_layer=nn.LayerNorm, downsample=None, use_checkpoint=False, img_size=224, patch_size=4, resi_connection="1conv", ): super(RSTB, self).__init__() self.dim = dim self.input_resolution = input_resolution self.residual_group = BasicLayer( dim=dim, input_resolution=input_resolution, depth=depth, num_heads=num_heads, window_size=window_size, mlp_ratio=mlp_ratio, qkv_bias=qkv_bias, qk_scale=qk_scale, drop=drop, attn_drop=attn_drop, drop_path=drop_path, norm_layer=norm_layer, downsample=downsample, use_checkpoint=use_checkpoint, ) if resi_connection == "1conv": self.conv = nn.Conv2d(dim, dim, 3, 1, 1) elif resi_connection == "3conv": # to save parameters and memory self.conv = nn.Sequential( nn.Conv2d(dim, dim // 4, 3, 1, 1), nn.LeakyReLU(negative_slope=0.2, inplace=True), nn.Conv2d(dim // 4, dim // 4, 1, 1, 0), nn.LeakyReLU(negative_slope=0.2, inplace=True), nn.Conv2d(dim // 4, dim, 3, 1, 1), ) self.patch_embed = PatchEmbed( img_size=img_size, patch_size=patch_size, in_chans=0, embed_dim=dim, norm_layer=None ) self.patch_unembed = PatchUnEmbed( img_size=img_size, patch_size=patch_size, in_chans=0, embed_dim=dim, norm_layer=None ) def forward(self, x, x_size): return ( self.patch_embed(self.conv(self.patch_unembed(self.residual_group(x, x_size), x_size))) + x ) def flops(self): flops = 0 flops += self.residual_group.flops() H, W = self.input_resolution flops += H * W * self.dim * self.dim * 9 flops += self.patch_embed.flops() flops += self.patch_unembed.flops() return flops class PatchEmbed(nn.Module): r"""Image to Patch Embedding Args: img_size (int): Image size. Default: 224. patch_size (int): Patch token size. Default: 4. in_chans (int): Number of input image channels. Default: 3. embed_dim (int): Number of linear projection output channels. Default: 96. norm_layer (nn.Module, optional): Normalization layer. Default: None """ def __init__(self, img_size=224, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None): super().__init__() img_size = to_2tuple(img_size) patch_size = to_2tuple(patch_size) patches_resolution = [img_size[0] // patch_size[0], img_size[1] // patch_size[1]] self.img_size = img_size self.patch_size = patch_size self.patches_resolution = patches_resolution self.num_patches = patches_resolution[0] * patches_resolution[1] self.in_chans = in_chans self.embed_dim = embed_dim if norm_layer is not None: self.norm = norm_layer(embed_dim) else: self.norm = None def forward(self, x): x = x.flatten(2).transpose(1, 2) # B Ph*Pw C if self.norm is not None: x = self.norm(x) return x def flops(self): flops = 0 H, W = self.img_size if self.norm is not None: flops += H * W * self.embed_dim return flops class PatchUnEmbed(nn.Module): r"""Image to Patch Unembedding Args: img_size (int): Image size. Default: 224. patch_size (int): Patch token size. Default: 4. in_chans (int): Number of input image channels. Default: 3. embed_dim (int): Number of linear projection output channels. Default: 96. norm_layer (nn.Module, optional): Normalization layer. Default: None """ def __init__(self, img_size=224, patch_size=4, in_chans=3, embed_dim=96, norm_layer=None): super().__init__() img_size = to_2tuple(img_size) patch_size = to_2tuple(patch_size) patches_resolution = [img_size[0] // patch_size[0], img_size[1] // patch_size[1]] self.img_size = img_size self.patch_size = patch_size self.patches_resolution = patches_resolution self.num_patches = patches_resolution[0] * patches_resolution[1] self.in_chans = in_chans self.embed_dim = embed_dim def forward(self, x, x_size): B, HW, C = x.shape x = x.transpose(1, 2).view(B, self.embed_dim, x_size[0], x_size[1]) # B Ph*Pw C return x def flops(self): flops = 0 return flops class Upsample(nn.Sequential): """Upsample module. Args: scale (int): Scale factor. Supported scales: 2^n and 3. num_feat (int): Channel number of intermediate features. """ def __init__(self, scale, num_feat): m = [] if (scale & (scale - 1)) == 0: # scale = 2^n for _ in range(int(math.log(scale, 2))): m.append(nn.Conv2d(num_feat, 4 * num_feat, 3, 1, 1)) m.append(nn.PixelShuffle(2)) elif scale == 3: m.append(nn.Conv2d(num_feat, 9 * num_feat, 3, 1, 1)) m.append(nn.PixelShuffle(3)) else: raise ValueError(f"scale {scale} is not supported. " "Supported scales: 2^n and 3.") super(Upsample, self).__init__(*m) class UpsampleOneStep(nn.Sequential): """UpsampleOneStep module (the difference with Upsample is that it always only has 1conv + 1pixelshuffle) Used in lightweight SR to save parameters. Args: scale (int): Scale factor. Supported scales: 2^n and 3. num_feat (int): Channel number of intermediate features. """ def __init__(self, scale, num_feat, num_out_ch, input_resolution=None): self.num_feat = num_feat self.input_resolution = input_resolution m = [] m.append(nn.Conv2d(num_feat, (scale ** 2) * num_out_ch, 3, 1, 1)) m.append(nn.PixelShuffle(scale)) super(UpsampleOneStep, self).__init__(*m) def flops(self): H, W = self.input_resolution flops = H * W * self.num_feat * 3 * 9 return flops class SwinIR(nn.Module): r"""SwinIR A PyTorch impl of : `SwinIR: Image Restoration Using Swin Transformer`, based on Swin Transformer. Args: img_size (int | tuple(int)): Input image size. Default 64 patch_size (int | tuple(int)): Patch size. Default: 1 in_chans (int): Number of input image channels. Default: 3 embed_dim (int): Patch embedding dimension. Default: 96 depths (tuple(int)): Depth of each Swin Transformer layer. num_heads (tuple(int)): Number of attention heads in different layers. window_size (int): Window size. Default: 7 mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4 qkv_bias (bool): If True, add a learnable bias to query, key, value. Default: True qk_scale (float): Override default qk scale of head_dim ** -0.5 if set. Default: None drop_rate (float): Dropout rate. Default: 0 attn_drop_rate (float): Attention dropout rate. Default: 0 drop_path_rate (float): Stochastic depth rate. Default: 0.1 norm_layer (nn.Module): Normalization layer. Default: nn.LayerNorm. ape (bool): If True, add absolute position embedding to the patch embedding. Default: False patch_norm (bool): If True, add normalization after patch embedding. Default: True use_checkpoint (bool): Whether to use checkpointing to save memory. Default: False upscale: Upscale factor. 2/3/4/8 for image SR, 1 for denoising and compress artifact reduction img_range: Image range. 1. or 255. upsampler: The reconstruction reconstruction module. 'pixelshuffle'/'pixelshuffledirect'/'nearest+conv'/None resi_connection: The convolutional block before residual connection. '1conv'/'3conv' """ def __init__( self, img_size=64, patch_size=1, in_chans=3, embed_dim=96, depths=[6, 6, 6, 6], num_heads=[6, 6, 6, 6], window_size=7, mlp_ratio=4.0, qkv_bias=True, qk_scale=None, drop_rate=0.0, attn_drop_rate=0.0, drop_path_rate=0.1, norm_layer=nn.LayerNorm, ape=False, patch_norm=True, use_checkpoint=False, upscale=2, img_range=1.0, upsampler="", resi_connection="1conv", **kwargs, ): super(SwinIR, self).__init__() num_in_ch = in_chans num_out_ch = in_chans num_feat = 64 self.img_range = img_range if in_chans == 3: rgb_mean = (0.4488, 0.4371, 0.4040) self.mean = flow.Tensor(rgb_mean).view(1, 3, 1, 1) else: self.mean = flow.zeros(1, 1, 1, 1) self.upscale = upscale self.upsampler = upsampler self.window_size = window_size # 1, shallow feature extraction self.conv_first = nn.Conv2d(num_in_ch, embed_dim, 3, 1, 1) # 2, deep feature extraction self.num_layers = len(depths) self.embed_dim = embed_dim self.ape = ape self.patch_norm = patch_norm self.num_features = embed_dim self.mlp_ratio = mlp_ratio # split image into non-overlapping patches self.patch_embed = PatchEmbed( img_size=img_size, patch_size=patch_size, in_chans=embed_dim, embed_dim=embed_dim, norm_layer=norm_layer if self.patch_norm else None, ) num_patches = self.patch_embed.num_patches patches_resolution = self.patch_embed.patches_resolution self.patches_resolution = patches_resolution # merge non-overlapping patches into image self.patch_unembed = PatchUnEmbed( img_size=img_size, patch_size=patch_size, in_chans=embed_dim, embed_dim=embed_dim, norm_layer=norm_layer if self.patch_norm else None, ) # absolute position embedding if self.ape: self.absolute_pos_embed = nn.Parameter(flow.zeros(1, num_patches, embed_dim)) trunc_normal_(self.absolute_pos_embed, std=0.02) self.pos_drop = nn.Dropout(p=drop_rate) # stochastic depth dpr = [ x.item() for x in flow.linspace(0, drop_path_rate, sum(depths)) ] # stochastic depth decay rule # build Residual Swin Transformer blocks (RSTB) self.layers = nn.ModuleList() for i_layer in range(self.num_layers): layer = RSTB( dim=embed_dim, input_resolution=(patches_resolution[0], patches_resolution[1]), depth=depths[i_layer], num_heads=num_heads[i_layer], window_size=window_size, mlp_ratio=self.mlp_ratio, qkv_bias=qkv_bias, qk_scale=qk_scale, drop=drop_rate, attn_drop=attn_drop_rate, drop_path=dpr[ sum(depths[:i_layer]) : sum(depths[: i_layer + 1]) ], # no impact on SR results norm_layer=norm_layer, downsample=None, use_checkpoint=use_checkpoint, img_size=img_size, patch_size=patch_size, resi_connection=resi_connection, ) self.layers.append(layer) self.norm = norm_layer(self.num_features) # build the last conv layer in deep feature extraction if resi_connection == "1conv": self.conv_after_body = nn.Conv2d(embed_dim, embed_dim, 3, 1, 1) elif resi_connection == "3conv": # to save parameters and memory self.conv_after_body = nn.Sequential( nn.Conv2d(embed_dim, embed_dim // 4, 3, 1, 1), nn.LeakyReLU(negative_slope=0.2, inplace=True), nn.Conv2d(embed_dim // 4, embed_dim // 4, 1, 1, 0), nn.LeakyReLU(negative_slope=0.2, inplace=True), nn.Conv2d(embed_dim // 4, embed_dim, 3, 1, 1), ) # 3, high quality image reconstruction if self.upsampler == "pixelshuffle": # for classical SR self.conv_before_upsample = nn.Sequential( nn.Conv2d(embed_dim, num_feat, 3, 1, 1), nn.LeakyReLU(inplace=True) ) self.upsample = Upsample(upscale, num_feat) self.conv_last = nn.Conv2d(num_feat, num_out_ch, 3, 1, 1) elif self.upsampler == "pixelshuffledirect": # for lightweight SR (to save parameters) self.upsample = UpsampleOneStep( upscale, embed_dim, num_out_ch, (patches_resolution[0], patches_resolution[1]) ) elif self.upsampler == "nearest+conv": # for real-world SR (less artifacts) self.conv_before_upsample = nn.Sequential( nn.Conv2d(embed_dim, num_feat, 3, 1, 1), nn.LeakyReLU(inplace=True) ) self.conv_up1 = nn.Conv2d(num_feat, num_feat, 3, 1, 1) if self.upscale == 4: self.conv_up2 = nn.Conv2d(num_feat, num_feat, 3, 1, 1) self.conv_hr = nn.Conv2d(num_feat, num_feat, 3, 1, 1) self.conv_last = nn.Conv2d(num_feat, num_out_ch, 3, 1, 1) self.lrelu = nn.LeakyReLU(negative_slope=0.2, inplace=True) else: # for image denoising and JPEG compression artifact reduction self.conv_last = nn.Conv2d(embed_dim, num_out_ch, 3, 1, 1) self.apply(self._init_weights) def _init_weights(self, m): if isinstance(m, nn.Linear): trunc_normal_(m.weight, std=0.02) if isinstance(m, nn.Linear) and m.bias is not None: nn.init.constant_(m.bias, 0) elif isinstance(m, nn.LayerNorm): nn.init.constant_(m.bias, 0) nn.init.constant_(m.weight, 1.0) def no_weight_decay(self): return {"absolute_pos_embed"} def no_weight_decay_keywords(self): return {"relative_position_bias_table"} def check_image_size(self, x): _, _, h, w = x.size() mod_pad_h = (self.window_size - h % self.window_size) % self.window_size mod_pad_w = (self.window_size - w % self.window_size) % self.window_size x = F.pad(x, (0, mod_pad_w, 0, mod_pad_h), "reflect") return x def forward_features(self, x): x_size = (x.shape[2], x.shape[3]) x = self.patch_embed(x) if self.ape: x = x + self.absolute_pos_embed x = self.pos_drop(x) for layer in self.layers: x = layer(x, x_size) x = self.norm(x) # B L C x = self.patch_unembed(x, x_size) return x def forward(self, x): H, W = x.shape[2:] x = self.check_image_size(x) self.mean = self.mean.type_as(x).to(x.device) x = (x - self.mean) * self.img_range if self.upsampler == "pixelshuffle": # for classical SR x = self.conv_first(x) x = self.conv_after_body(self.forward_features(x)) + x x = self.conv_before_upsample(x) x = self.conv_last(self.upsample(x)) elif self.upsampler == "pixelshuffledirect": # for lightweight SR x = self.conv_first(x) x = self.conv_after_body(self.forward_features(x)) + x x = self.upsample(x) elif self.upsampler == "nearest+conv": # for real-world SR x = self.conv_first(x) x = self.conv_after_body(self.forward_features(x)) + x x = self.conv_before_upsample(x) x = self.lrelu( self.conv_up1(flow.nn.functional.interpolate(x, scale_factor=2, mode="nearest")) ) if self.upscale == 4: x = self.lrelu( self.conv_up2(flow.nn.functional.interpolate(x, scale_factor=2, mode="nearest")) ) x = self.conv_last(self.lrelu(self.conv_hr(x))) else: # for image denoising and JPEG compression artifact reduction x_first = self.conv_first(x) res = self.conv_after_body(self.forward_features(x_first)) + x_first x = x + self.conv_last(res) x = x / self.img_range + self.mean return x[:, :, : H * self.upscale, : W * self.upscale] def flops(self): flops = 0 H, W = self.patches_resolution flops += H * W * 3 * self.embed_dim * 9 flops += self.patch_embed.flops() for i, layer in enumerate(self.layers): flops += layer.flops() flops += H * W * 3 * self.embed_dim * self.embed_dim flops += self.upsample.flops() return flops if __name__ == "__main__": upscale = 4 window_size = 8 height = (1024 // upscale // window_size + 1) * window_size width = (720 // upscale // window_size + 1) * window_size model = SwinIR( upscale=2, img_size=(height, width), window_size=window_size, img_range=1.0, depths=[6, 6, 6, 6], embed_dim=60, num_heads=[6, 6, 6, 6], mlp_ratio=2, upsampler="pixelshuffledirect", ) print(model) print(height, width, model.flops() / 1e9) x = flow.randn((1, 3, height, width)) x = model(x) print(x.shape)
37,112
34.823359
100
py
libai
libai-main/projects/DALLE2/swinir/__init__.py
from .models import SwinIR from .upsample import load_model, upsample4x, upsample16x
85
27.666667
57
py
libai
libai-main/projects/DALLE2/swinir/upsample.py
import os import oneflow as flow import requests from .models import SwinIR as net def load_torch_weight(model, model_path): # load torch weight import torch param_key_g = "params_ema" pretrained_model = torch.load(model_path, map_location="cpu") pretrained_model = ( pretrained_model[param_key_g] if param_key_g in pretrained_model.keys() else pretrained_model ) new_state_dict = {} for k, v in pretrained_model.items(): flow_tensor = flow.tensor(v.numpy()) new_state_dict[k] = flow_tensor model.load_state_dict(new_state_dict, strict=True) return model def load_model(model_path=None): # set up model if os.path.exists(model_path): print(f"loading model from {model_path}") else: os.makedirs(os.path.dirname(model_path), exist_ok=True) url = "https://github.com/JingyunLiang/SwinIR/releases/download/v0.0/{}".format( os.path.basename(model_path) ) r = requests.get(url, allow_redirects=True) print(f"downloading model {model_path}") open(model_path, "wb").write(r.content) model = net( upscale=4, in_chans=3, img_size=64, window_size=8, img_range=1.0, depths=[6, 6, 6, 6, 6, 6, 6, 6, 6], embed_dim=240, num_heads=[8, 8, 8, 8, 8, 8, 8, 8, 8], mlp_ratio=2, upsampler="nearest+conv", resi_connection="3conv", ) model = load_torch_weight(model, model_path) return model def upsample4x(img_lq, model): """upsample img from h*w to (4h) * (4w)""" device = flow.device("cuda" if flow.cuda.is_available() else "cpu") model.eval() model = model.to(device) img_lq = img_lq.to(device) window_size = 8 scale = 4 # inference with flow.no_grad(): # pad input image to be a multiple of window_size _, _, h_old, w_old = img_lq.size() h_pad = (h_old // window_size + 1) * window_size - h_old w_pad = (w_old // window_size + 1) * window_size - w_old img_lq = flow.cat([img_lq, flow.flip(img_lq, [2])], 2)[:, :, : h_old + h_pad, :] img_lq = flow.cat([img_lq, flow.flip(img_lq, [3])], 3)[:, :, :, : w_old + w_pad] output = model(img_lq) output = output[..., : h_old * scale, : w_old * scale] output = output.clamp_(0, 1) return output def upsample16x(imgs, model): return upsample4x(upsample4x(imgs, model), model)
2,486
28.607143
88
py
libai
libai-main/projects/BLOOM/configs/bloom_inference.py
from omegaconf import DictConfig from libai.config import LazyCall from projects.BLOOM.modeling.bloom_model import BloomModel cfg = dict( # model vocab_size=250880, hidden_size=64, hidden_layers=2, n_head=8, padding_idx=3, layer_norm_epsilon=1e-5, initializer_range=0.02, apply_residual_connection_post_layernorm=False, hidden_dropout=0.0, attention_dropout=0.0, pretraining_tp=1, slow_but_exact=False, amp_enabled=False, # Inference is_encoder_decoder=False, max_length=512, min_length=0, do_sample=False, early_stopping=False, num_beams=1, num_beam_groups=1, diversity_penalty=0.0, temperature=1.0, top_k=50, top_p=1.0, typical_p=1.0, repetition_penalty=1.0, length_penalty=1.0, no_repeat_ngram_size=0, encoder_no_repeat_ngram_size=0, num_return_sequences=1, chunk_size_feed_forward=0, output_scores=False, forced_bos_token_id=None, forced_eos_token_id=None, remove_invalid_values=False, exponential_decay_length_penalty=None, use_cache=True, # Tokenizer pad_token_id=3, eos_token_id=2, bos_token_id=1, sep_token_id=None, decoder_start_token_id=None, ) cfg = DictConfig(cfg) glm_model = LazyCall(BloomModel)(cfg=cfg)
1,303
21.877193
58
py
libai
libai-main/projects/BLOOM/utils/model_loader.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import json from libai.models.utils import ModelLoaderHuggerFace, ModelLoaderLiBai class BlooMLoaderHuggerFace(ModelLoaderHuggerFace): def __init__(self, model, libai_cfg, pretrained_model_path, **kwargs): super().__init__(model, libai_cfg, pretrained_model_path, **kwargs) """NOTE: base_model_prefix_1 is BLOOM's prefix in Transformers. base_model_prefix_2 is BLOOM's prefix in LiBai.""" self.base_model_prefix_1 = "transformer" self.base_model_prefix_2 = "transformer" def _convert_state_dict(self, flow_state_dict, cfg): """Convert state_dict's keys to match model. Args: flow_state_dict (OrderedDict): model state dict. cfg (dict): model's default config dict in LiBai. Returns: OrderedDict: flow state dict. """ # The converted checkpoint. oneflow_state_dict = flow_state_dict.copy() old_keys = list(oneflow_state_dict.keys()) # prefix has_prefix = any(s.startswith(self.base_model_prefix_1) for s in oneflow_state_dict) prefix2 = "transformer." if has_prefix else "" # Convert layers. for key in old_keys: oneflow_state_dict[prefix2 + key] = oneflow_state_dict.pop(key) return oneflow_state_dict def _load_config_from_json(self, config_file): """load config from `config.json`, and update default config. Args: config_file (str): Path of config file. """ with open(config_file, mode="r", encoding="utf-8") as f: cfg_dict = json.load(f) self._update_cfg("hidden_layers", cfg_dict["n_layer"]) self._update_cfg("hidden_size", cfg_dict["n_embed"]) self._update_cfg("n_head", cfg_dict["num_attention_heads"]) # update libai_cfg by config.json for k, v in cfg_dict.items(): self._update_cfg(k, v) # update libai_cfg by kwargs for k, v in self.kwargs.items(): self._update_cfg(k, v) self._update_cfg_log() class BlooMLoaderLibai(ModelLoaderLiBai): def __init__(self, model, libai_cfg, pretrained_model_path, **kwargs): super().__init__(model, libai_cfg, pretrained_model_path, **kwargs) self.base_model_prefix_2 = "transformer"
2,934
34.792683
92
py
libai
libai-main/projects/BLOOM/modeling/activation.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # Copyright 2022 The HuggingFace Inc. team. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow from oneflow import nn def bloom_gelu_forward(x): """ Custom bias GELU function. Adapted from Megatron-DeepSpeed code. Here we use a simple implementation (inference) to make the model jitable. Args: x (`torch.tensor`, *required*): input hidden states """ return x * 0.5 * (1.0 + flow.tanh(0.79788456 * x * (1 + 0.044715 * x * x))) def bloom_gelu_back(g, x): """ gradient of tanh approximation of gelu gradient of actual gelu is: 0.5 * (1. + torch.erf(x * 0.70710678)) + 0.3989423 * x * torch.exp(-0.5 * x * x) Args: g (`torch.tensor`, *required*): gradient output tensor x (`torch.tensor`, *required*): input tensor """ x = x[0] tanh_out = flow.tanh(0.79788456 * x * (1 + 0.044715 * x * x)) ff = 0.5 * x * ((1 - tanh_out * tanh_out) * (0.79788456 + 0.1070322243 * x * x)) + 0.5 * ( 1 + tanh_out ) return ff * g class GeLUFunction(flow.autograd.Function): @staticmethod def forward(ctx, input): ctx.save_for_backward(input) return bloom_gelu_forward(input) @staticmethod def backward(ctx, grad_output): input = ctx.saved_tensors tmp = bloom_gelu_back(grad_output, input) return tmp class BloomGelu(nn.Module): """ BloomBiasGelu wrapper function that make use of the simple function on inference mode to make the model torchscriptable and use the autograd function in training mode to get the accurate results of the gradients Partly copied from Megatron-DeepSpeed code and adapted for our needs See here why autograd functions are not torchscriptable: https://github.com/pytorch/pytorch/issues/22329 """ def __init__(self): super().__init__() def forward(self, x): if self.training: return GeLUFunction.apply(x) else: return bloom_gelu_forward(x)
2,621
30.590361
97
py
libai
libai-main/projects/BLOOM/modeling/transformers.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # Copyright 2022 The HuggingFace Inc. team. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from oneflow import nn from libai.layers import LayerNorm from libai.utils import distributed as dist from projects.BLOOM.modeling.attention import BloomAttention from projects.BLOOM.modeling.mlp import BloomMLP class BloomBlock(nn.Module): def __init__( self, hidden_size, n_head, layer_norm_epsilon, hidden_dropout, attention_dropout, pretraining_tp, slow_but_exact, init_method, output_layer_init_method, apply_residual_connection_post_layernorm, layer_idx=0, ): super().__init__() hidden_size = hidden_size self.input_layernorm = LayerNorm(hidden_size, eps=layer_norm_epsilon, layer_idx=layer_idx) self.num_heads = n_head self.self_attention = BloomAttention( hidden_size=hidden_size, n_head=n_head, hidden_dropout=hidden_dropout, attention_dropout=attention_dropout, pretraining_tp=pretraining_tp, slow_but_exact=slow_but_exact, init_method=init_method, output_layer_init_method=output_layer_init_method, layer_idx=layer_idx, ) self.post_attention_layernorm = LayerNorm( hidden_size, eps=layer_norm_epsilon, layer_idx=layer_idx ) self.mlp = BloomMLP( hidden_size, pretraining_tp, slow_but_exact, hidden_dropout, init_method, output_layer_init_method, layer_idx, ) self.apply_residual_connection_post_layernorm = apply_residual_connection_post_layernorm self.hidden_dropout = hidden_dropout self.layer_idx = layer_idx def forward( self, hidden_states, alibi, attention_mask, layer_past=None, head_mask=None, use_cache: bool = False, output_attentions: bool = False, ): # Change placement for pipeline parallelsim hidden_states = hidden_states.to_global(placement=dist.get_layer_placement(self.layer_idx)) alibi = alibi.to_global(placement=dist.get_layer_placement(self.layer_idx)) # hidden_states shape: (batch_size, seq_length, hidden_size) if attention_mask is not None: attention_mask = attention_mask.to_global( placement=dist.get_layer_placement(self.layer_idx) ) layernorm_output = self.input_layernorm(hidden_states) if self.apply_residual_connection_post_layernorm: residual = layernorm_output else: residual = hidden_states # Self attention. attn_outputs = self.self_attention( layernorm_output, residual, layer_past=layer_past, attention_mask=attention_mask, alibi=alibi, head_mask=head_mask, use_cache=use_cache, output_attentions=output_attentions, ) attention_output = attn_outputs[0] outputs = attn_outputs[1:] layernorm_output = self.post_attention_layernorm(attention_output) if self.apply_residual_connection_post_layernorm: residual = layernorm_output else: residual = attention_output # MLP. output = self.mlp(layernorm_output, residual) if use_cache: outputs = (output,) + outputs else: outputs = (output,) + outputs[1:] return outputs # hidden_states, present, attentions
4,242
30.664179
99
py
libai
libai-main/projects/BLOOM/modeling/mlp.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # Copyright 2022 The HuggingFace Inc. team. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow import oneflow.nn.functional as F from oneflow import nn from libai.layers import Linear from projects.BLOOM.modeling.activation import BloomGelu from projects.BLOOM.modeling.attention import dropout_add class BloomMLP(nn.Module): def __init__( self, hidden_size, pretraining_tp, slow_but_exact, hidden_dropout, init_method=None, output_layer_init_method=None, layer_idx=0, ): super().__init__() hidden_size = hidden_size if output_layer_init_method is None: output_layer_init_method = init_method self.pretraining_tp = pretraining_tp self.slow_but_exact = slow_but_exact self.dense_h_to_4h = Linear( hidden_size, 4 * hidden_size, parallel="col", init_method=init_method, layer_idx=layer_idx, ) self.gelu_impl = BloomGelu() self.dense_4h_to_h = Linear( 4 * hidden_size, hidden_size, parallel="row", init_method=output_layer_init_method, layer_idx=layer_idx, ) self.hidden_dropout = hidden_dropout def forward(self, hidden_states, residual): hidden_states = self.gelu_impl(self.dense_h_to_4h(hidden_states)) if self.pretraining_tp > 1 and self.slow_but_exact: intermediate_output = flow.zeros_like(residual) slices = self.dense_4h_to_h.weight.shape[-1] / self.pretraining_tp for i in range(self.pretraining_tp): intermediate_output = intermediate_output + F.linear( hidden_states[:, :, int(i * slices) : int((i + 1) * slices)], self.dense_4h_to_h.weight[:, int(i * slices) : int((i + 1) * slices)], ) else: intermediate_output = self.dense_4h_to_h(hidden_states) output = dropout_add(intermediate_output, residual, self.hidden_dropout, self.training) return output
2,716
33.833333
95
py
libai
libai-main/projects/BLOOM/modeling/bloom_model.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # Copyright 2022 The HuggingFace Inc. team. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow from oneflow import nn from libai.config import configurable from libai.inference.generator.generation_utils import Generator from libai.layers import Embedding, LayerNorm, LMLogits from libai.models.utils import init_method_normal, scaled_init_method_normal from libai.utils import distributed as dist from projects.BLOOM.modeling.mask import _expand_mask, _make_causal_mask, build_alibi_tensor from projects.BLOOM.modeling.transformers import BloomBlock class BloomModel(nn.Module): @configurable def __init__( self, vocab_size, hidden_size, hidden_layers, n_head, padding_idx, pretraining_tp=1, slow_but_exact=False, initializer_range=0.02, apply_residual_connection_post_layernorm=False, hidden_dropout=0, attention_dropout=0, amp_enabled=False, layer_norm_epsilon=1e-12, cfg=None, ): super().__init__() self.cfg = cfg self.embed_dim = hidden_size self.num_heads = n_head self.hidden_layers = hidden_layers init_method = init_method_normal(initializer_range) scaled_init_method = scaled_init_method_normal(initializer_range, hidden_layers) self.word_embeddings = Embedding( vocab_size, self.embed_dim, padding_idx=padding_idx, init_method=init_method, amp_enabled=amp_enabled, layer_idx=0, ) self.word_embeddings_layernorm = LayerNorm( self.embed_dim, eps=layer_norm_epsilon, layer_idx=0 ) self.h = flow.nn.ModuleList( [ BloomBlock( hidden_size=hidden_size, n_head=n_head, layer_norm_epsilon=layer_norm_epsilon, hidden_dropout=hidden_dropout, attention_dropout=attention_dropout, pretraining_tp=pretraining_tp, slow_but_exact=slow_but_exact, init_method=init_method, output_layer_init_method=scaled_init_method, apply_residual_connection_post_layernorm=apply_residual_connection_post_layernorm, # noqa layer_idx=i, ) for i in range(hidden_layers) ] ) # Final Layer Norm self.ln_f = LayerNorm(self.embed_dim, eps=layer_norm_epsilon, layer_idx=hidden_layers - 1) @classmethod def from_config(cls, cfg): return { "vocab_size": cfg.vocab_size, "hidden_size": cfg.hidden_size, "hidden_layers": cfg.hidden_layers, "n_head": cfg.n_head, "padding_idx": cfg.padding_idx, "pretraining_tp": cfg.pretraining_tp, "slow_but_exact": cfg.slow_but_exact, "apply_residual_connection_post_layernorm": cfg.apply_residual_connection_post_layernorm, # noqa "hidden_dropout": cfg.hidden_dropout, "attention_dropout": cfg.attention_dropout, "amp_enabled": cfg.amp_enabled, "layer_norm_epsilon": cfg.layer_norm_epsilon, "cfg": cfg, } def _prepare_attn_mask( self, attention_mask, input_shape, past_key_values_length, ): combined_attention_mask = None _, src_length = input_shape if src_length > 1: combined_attention_mask = _make_causal_mask( input_shape, past_key_values_length=past_key_values_length ) # [batch_size, seq_length] -> [batch_size, 1, tgt_length, src_length] expanded_attn_mask = _expand_mask(attention_mask, tgt_length=src_length) combined_attention_mask = ( expanded_attn_mask if combined_attention_mask is None else expanded_attn_mask | combined_attention_mask ) return combined_attention_mask def _convert_head_mask_to_5d(self, head_mask, num_hidden_layers): """-> [num_hidden_layers x batch x num_heads x seq_length x seq_length]""" if head_mask.dim() == 1: head_mask = head_mask.unsqueeze(0).unsqueeze(0).unsqueeze(-1).unsqueeze(-1) head_mask = head_mask.expand(num_hidden_layers, -1, -1, -1, -1) elif head_mask.dim() == 2: head_mask = ( head_mask.unsqueeze(1).unsqueeze(-1).unsqueeze(-1) ) # We can specify head_mask for each layer assert head_mask.dim() == 5, f"head_mask.dim != 5, instead {head_mask.dim()}" head_mask = head_mask.to(dtype=self.dtype) # switch to float if need + fp16 compatibility return head_mask def get_head_mask(self, head_mask, num_hidden_layers, is_attention_chunked=False): """ Prepare the head mask if needed. Args: head_mask (`torch.Tensor` with shape `[num_heads]` or `[num_hidden_layers x num_heads]`, *optional*): The mask indicating if we should keep the heads or not (1.0 for keep, 0.0 for discard). num_hidden_layers (`int`): The number of hidden layers in the model. is_attention_chunked: (`bool`, *optional*, defaults to `False`): Whether or not the attentions scores are computed by chunks or not. Returns: `torch.Tensor` with shape `[num_hidden_layers x batch x num_heads x seq_length x seq_length]` or list with `[None]` for each layer. """ if head_mask is not None: head_mask = self._convert_head_mask_to_5d(head_mask, num_hidden_layers) if is_attention_chunked is True: head_mask = head_mask.unsqueeze(-1) else: head_mask = [None] * num_hidden_layers return head_mask def forward( self, input_ids=None, attention_mask=None, past_key_values=None, head_mask=None, inputs_embeds=None, use_cache=None, output_attentions=None, ): input_ids = ( input_ids.to_global(placement=dist.get_layer_placement(0)) if input_ids is not None else input_ids ) attention_mask = ( attention_mask.to_global(placement=dist.get_layer_placement(0)) if attention_mask is not None else attention_mask ) head_mask = ( head_mask.to_global(placement=dist.get_layer_placement(0)) if head_mask is not None else head_mask ) inputs_embeds = ( inputs_embeds.to_global(placement=dist.get_layer_placement(0)) if inputs_embeds is not None else inputs_embeds ) if input_ids is not None: batch_size, seq_length = input_ids.shape if past_key_values is None: past_key_values = tuple([None] * len(self.h)) head_mask = self.get_head_mask(head_mask, self.hidden_layers) if inputs_embeds is None: inputs_embeds = self.word_embeddings(input_ids) hidden_states = self.word_embeddings_layernorm(inputs_embeds) presents = () if use_cache else None seq_length_with_past = seq_length past_key_values_length = 0 if past_key_values[0] is not None: past_key_values_length = past_key_values[0][0].shape[2] seq_length_with_past = seq_length_with_past + past_key_values_length if attention_mask is None: attention_mask = flow.ones( (batch_size, seq_length_with_past), sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) alibi = build_alibi_tensor(attention_mask, self.num_heads, hidden_states.dtype) causal_mask = self._prepare_attn_mask( attention_mask, input_shape=(batch_size, seq_length), past_key_values_length=past_key_values_length, ) for i, (block, layer_past) in enumerate(zip(self.h, past_key_values)): outputs = block( hidden_states, layer_past=layer_past, attention_mask=causal_mask, head_mask=head_mask[i], use_cache=use_cache, output_attentions=output_attentions, alibi=alibi, ) hidden_states = outputs[0] if use_cache is True: presents = presents + (outputs[1],) hidden_states = self.ln_f(hidden_states) return {"last_hidden_state": hidden_states, "past_key_values": presents} class BloomForCausalLM(nn.Module, Generator): @configurable def __init__( self, vocab_size, hidden_size, hidden_layers, n_head, padding_idx, pretraining_tp=1, slow_but_exact=False, initializer_range=0.02, apply_residual_connection_post_layernorm=False, hidden_dropout=0, attention_dropout=0, amp_enabled=False, layer_norm_epsilon=1e-12, cfg=None, ): super().__init__() self.cfg = cfg self.transformer = BloomModel( vocab_size=vocab_size, hidden_size=hidden_size, hidden_layers=hidden_layers, n_head=n_head, padding_idx=padding_idx, pretraining_tp=pretraining_tp, slow_but_exact=slow_but_exact, initializer_range=initializer_range, apply_residual_connection_post_layernorm=apply_residual_connection_post_layernorm, hidden_dropout=hidden_dropout, attention_dropout=attention_dropout, amp_enabled=amp_enabled, layer_norm_epsilon=layer_norm_epsilon, cfg=cfg, ) self.lm_head = LMLogits(vocab_size, bias=False) @classmethod def from_config(cls, cfg): return { "vocab_size": cfg.vocab_size, "hidden_size": cfg.hidden_size, "hidden_layers": cfg.hidden_layers, "n_head": cfg.n_head, "padding_idx": cfg.padding_idx, "pretraining_tp": cfg.pretraining_tp, "slow_but_exact": cfg.slow_but_exact, "apply_residual_connection_post_layernorm": cfg.apply_residual_connection_post_layernorm, # noqa "hidden_dropout": cfg.hidden_dropout, "attention_dropout": cfg.attention_dropout, "amp_enabled": cfg.amp_enabled, "layer_norm_epsilon": cfg.layer_norm_epsilon, "cfg": cfg, } def forward( self, input_ids=None, attention_mask=None, past_key_values=None, head_mask=None, inputs_embeds=None, use_cache=None, output_attentions=None, ): transformer_outputs = self.transformer( input_ids, past_key_values=past_key_values, attention_mask=attention_mask, head_mask=head_mask, inputs_embeds=inputs_embeds, use_cache=use_cache, output_attentions=output_attentions, ) hidden_states = transformer_outputs["last_hidden_state"] lm_logits = self.lm_head(hidden_states, self.transformer.word_embeddings.weight) return { "logits": lm_logits, "past_key_values": transformer_outputs["past_key_values"], "hidden_states": transformer_outputs["last_hidden_state"], # "attentions": transformer_outputs.attentions, } def prepare_inputs_for_generation( self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs, ) -> dict: # only last token for input_ids if past is not None if past_key_values: input_ids = input_ids[:, -1].unsqueeze(-1) if past_key_values[0][0].shape[0] == input_ids.shape[0]: past_key_values = self._convert_to_bloom_cache(past_key_values) if inputs_embeds is not None and past_key_values is None: model_inputs = {"inputs_embeds": inputs_embeds} else: model_inputs = {"input_ids": input_ids} model_inputs.update( { "past_key_values": past_key_values, "use_cache": kwargs.get("use_cache"), "attention_mask": attention_mask, } ) return model_inputs def _reorder_cache(self, past, beam_idx): standardized_past = self._convert_to_standard_cache(past, batch_size=len(beam_idx)) device_to_beam_idx = { past_state.device: beam_idx.to(past_state.device) for layer_past in past for past_state in layer_past } reordered_past = tuple( ( layer_past[0].index_select(0, device_to_beam_idx[layer_past[0].device]), layer_past[1].index_select(0, device_to_beam_idx[layer_past[0].device]), ) for layer_past in standardized_past ) return self._convert_to_bloom_cache(reordered_past) def _convert_to_standard_cache( past_key_value, batch_size, ): """ Standardizes the format of the cache so as to match most implementations, i.e. to tuple(tuple([batch_size, num_heads, ...])) """ batch_size_times_num_heads, head_dim, seq_length = past_key_value[0][0].shape num_heads = batch_size_times_num_heads // batch_size return tuple( ( layer_past[0].view(batch_size, num_heads, head_dim, seq_length), layer_past[1].view(batch_size, num_heads, seq_length, head_dim), ) for layer_past in past_key_value ) def _convert_to_bloom_cache(past_key_value): """ Converts the cache to the format expected by Bloom, i.e. to tuple(tuple([batch_size * num_heads, ...])) """ batch_size, num_heads, head_dim, seq_length = past_key_value[0][0].shape batch_size_times_num_heads = batch_size * num_heads return tuple( ( layer_past[0].view(batch_size_times_num_heads, head_dim, seq_length), layer_past[1].view(batch_size_times_num_heads, seq_length, head_dim), ) for layer_past in past_key_value )
15,297
35.080189
110
py
libai
libai-main/projects/BLOOM/modeling/attention.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # Copyright 2022 The HuggingFace Inc. team. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import math import oneflow as flow from oneflow import nn from oneflow.nn import functional as F from libai.layers import Linear def dropout_add(x, residual, prob, training): """ Dropout add function Args: x (`torch.tensor`, *required*): input tensor residual (`torch.tensor`, *required*): esidual tensor prob (`float`, *required*): dropout probability training (`bool`, *required*): training mode """ out = F.dropout(x, p=prob, training=training) out = residual + out return out class BloomAttention(nn.Module): def __init__( self, hidden_size, n_head, hidden_dropout, attention_dropout, pretraining_tp, slow_but_exact, init_method=nn.init.xavier_normal_, output_layer_init_method=None, layer_idx=0, ): super().__init__() self.pretraining_tp = pretraining_tp self.slow_but_exact = slow_but_exact self.hidden_size = hidden_size self.num_heads = n_head self.head_dim = self.hidden_size // self.num_heads self.split_size = self.hidden_size self.hidden_dropout = hidden_dropout if output_layer_init_method is None: output_layer_init_method = init_method if self.head_dim * self.num_heads != self.hidden_size: raise ValueError( f"`hidden_size` must be divisible by num_heads " f"(got `hidden_size`: {self.hidden_size} and `num_heads`:" f" {self.num_heads})." ) # Layer-wise attention scaling self.inv_norm_factor = 1.0 / math.sqrt(self.head_dim) self.beta = 1.0 self.query_key_value = Linear( self.hidden_size, 3 * self.hidden_size, bias=True, parallel="col", init_method=init_method, layer_idx=layer_idx, ) self.dense = Linear( self.hidden_size, self.hidden_size, parallel="row", init_method=output_layer_init_method, layer_idx=layer_idx, ) self.attention_dropout = nn.Dropout(attention_dropout) def _split_heads(self, fused_qkv): """ Split the last dimension into (num_heads, head_dim) without making any copies, results share same memory storage as `fused_qkv` Args: fused_qkv (`torch.tensor`, *required*): [batch_size, seq_length, num_heads * 3 * head_dim] Returns: query: [batch_size, seq_length, num_heads, head_dim] key: [batch_size, seq_length, num_heads, head_dim] value: [batch_size, seq_length, num_heads, head_dim] """ batch_size, seq_length, three_times_hidden_size = fused_qkv.shape fused_qkv = fused_qkv.view(batch_size, seq_length, self.num_heads, 3, self.head_dim) return fused_qkv[..., 0, :], fused_qkv[..., 1, :], fused_qkv[..., 2, :] def _merge_heads(self, x): """ Merge heads together over the last dimenstion Args: x: (`torch.tensor`, *required*): [batch_size * num_heads, seq_length, head_dim] Returns: torch.tensor: [batch_size, seq_length, num_heads * head_dim] """ # What we want to achieve is: # batch_size * num_heads, seq_len, head_dim -> batch_size, seq_len, num_heads * head_dim batch_size_and_num_heads, seq_length, _ = x.shape batch_size = batch_size_and_num_heads // self.num_heads # First view to decompose the batch size # batch_size * num_heads, seq_len, head_dim -> batch_size, num_heads, seq_len, head_dim x = x.view(batch_size, self.num_heads, seq_length, self.head_dim) # batch_size, num_heads, seq_length, head_dim -> batch_size, seq_length, num_heads, head_dim x = x.permute(0, 2, 1, 3) # batch_size, seq_len, num_heads, head_dim -> batch_size, seq_len, num_heads * head_dim return x.reshape(batch_size, seq_length, self.num_heads * self.head_dim) def forward( self, hidden_states, residual, alibi, attention_mask, layer_past=None, head_mask=None, use_cache=False, output_attentions=False, ): fused_qkv = self.query_key_value(hidden_states) # [batch_size, seq_length, 3 x hidden_size] # 3 x [batch_size, seq_length, num_heads, head_dim] (query_layer, key_layer, value_layer) = self._split_heads(fused_qkv) batch_size, q_length, _, _ = query_layer.shape query_layer = query_layer.transpose(1, 2).reshape( batch_size * self.num_heads, q_length, self.head_dim ) key_layer = key_layer.permute(0, 2, 3, 1).reshape( batch_size * self.num_heads, self.head_dim, q_length ) value_layer = value_layer.transpose(1, 2).reshape( batch_size * self.num_heads, q_length, self.head_dim ) if layer_past is not None: past_key, past_value = layer_past key_layer = flow.cat((past_key, key_layer), dim=2) value_layer = flow.cat((past_value, value_layer), dim=1) _, _, kv_length = key_layer.shape if use_cache is True: present = (key_layer, value_layer) else: present = None matmul_result = flow.baddbmm( alibi, batch1=query_layer, batch2=key_layer, beta=self.beta, alpha=self.inv_norm_factor, ) attention_scores = matmul_result.view(batch_size, self.num_heads, q_length, kv_length) input_dtype = attention_scores.dtype attn_weights = flow.masked_fill( attention_scores, attention_mask, flow.finfo(attention_scores.dtype).min ) attention_probs = F.softmax(attn_weights, dim=-1).to(input_dtype) attention_probs = self.attention_dropout(attention_probs) if head_mask is not None: attention_probs = attention_probs * head_mask attention_probs_reshaped = attention_probs.view( batch_size * self.num_heads, q_length, kv_length ) context_layer = flow.bmm(attention_probs_reshaped, value_layer) context_layer = self._merge_heads(context_layer) if self.pretraining_tp > 1 and self.slow_but_exact: slices = self.hidden_size / self.pretraining_tp output_tensor = flow.zeros_like(context_layer) for i in range(self.pretraining_tp): output_tensor = output_tensor + F.linear( context_layer[:, :, int(i * slices) : int((i + 1) * slices)], self.dense.weight[:, int(i * slices) : int((i + 1) * slices)], ) else: output_tensor = self.dense(context_layer) output_tensor = dropout_add(output_tensor, residual, self.hidden_dropout, self.training) outputs = (output_tensor, present) if output_attentions: outputs += (attention_probs,) return outputs
7,869
33.669604
100
py
libai
libai-main/projects/BLOOM/modeling/mask.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # Copyright 2022 The HuggingFace Inc. team. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import math import oneflow as flow from libai.utils import distributed as dist def _make_causal_mask(input_ids_shape, past_key_values_length): """ Make causal mask used for self-attention. """ batch_size, target_length = input_ids_shape mask = flow.ones( (target_length, target_length + past_key_values_length), dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) # ONNX doesn't support `torch.Tensor.triu` properly, thus we use this workaround seq_ids = flow.arange( target_length, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) mask[:, past_key_values_length:] = seq_ids[:, None] < seq_ids[None, :] if past_key_values_length > 0: mask[:, :past_key_values_length] = False expanded_mask = mask[None, None, :, :].expand( batch_size, 1, target_length, target_length + past_key_values_length ) return expanded_mask def _expand_mask(mask, tgt_length): """ Expands attention_mask from `[batch_size, src_length]` to `[batch_size, 1, tgt_length, src_length]`. """ batch_size, src_length = mask.shape tgt_length = tgt_length if tgt_length is not None else src_length expanded_mask = ~(mask[:, None, None, :].to(flow.bool)) return expanded_mask.expand(batch_size, 1, tgt_length, src_length) def build_alibi_tensor(attention_mask, num_heads, dtype): batch_size, seq_length = attention_mask.shape closest_power_of_2 = 2 ** math.floor(math.log2(num_heads)) base = flow.tensor( 2 ** (-(2 ** -(math.log2(closest_power_of_2) - 3))), sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=attention_mask.placement, ) powers = flow.arange( 1, 1 + closest_power_of_2, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=attention_mask.placement, ) slopes = flow.pow(base, powers) if closest_power_of_2 != num_heads: extra_base = flow.tensor( 2 ** (-(2 ** -(math.log2(2 * closest_power_of_2) - 3))), sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=attention_mask.placement, ) num_remaining_heads = min(closest_power_of_2, num_heads - closest_power_of_2) extra_powers = flow.arange( 1, 1 + 2 * num_remaining_heads, 2, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=attention_mask.placement, ) slopes = flow.cat([slopes, flow.pow(extra_base, extra_powers)], dim=0) arange_tensor = ((attention_mask.cumsum(dim=-1) - 1) * attention_mask)[:, None, :] alibi = slopes[..., None] * arange_tensor return alibi.reshape(batch_size * num_heads, 1, seq_length).to(dtype)
3,662
35.63
86
py
libai
libai-main/projects/T5/models/embedding.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow import oneflow.nn as nn from oneflow.nn import init import libai.utils.distributed as dist from libai.layers.embedding import VocabEmbedding class T5Embedding(flow.nn.Module): def __init__( self, hidden_size, vocab_size, embedding_dropout_prob, init_method=flow.nn.init.xavier_normal_, amp_enabled=False, ) -> None: super().__init__() self.hidden_size = hidden_size self.vocab_size = vocab_size self.word_embeddings = VocabEmbedding( num_embeddings=vocab_size, embedding_dim=hidden_size, init_method=init_method, amp_enabled=amp_enabled, ) self.embedding_dropout = flow.nn.Dropout(embedding_dropout_prob) def forward(self, input_ids): word_embeddings = self.word_embeddings(input_ids) embeddings = self.embedding_dropout(word_embeddings) return embeddings class Embedding(nn.Module): """Construct the trainable embedding module, which does not support parallelization. This can be used for positional embedding and token type embedding. Arguments: num_embeddings: size of vocabulary. embedding_dim: dimension of embeddings. padding_idx: pad index. Defaults to None. init_method: method to initialize weights. Defaults to ``flow.nn.init.xavier_normal_``. amp_enabled: fp16 option for embedding weight. Defaults to False. """ def __init__( self, num_embeddings, embedding_dim, padding_idx=None, init_method=init.xavier_normal_, amp_enabled=False, layer_idx=0, ): super().__init__() self.num_embeddings = num_embeddings self.embedding_dim = embedding_dim if padding_idx is not None: if padding_idx > 0: assert ( padding_idx < self.num_embeddings ), "Padding_idx must be within num_embeddings" elif padding_idx < 0: assert ( padding_idx >= -self.num_embeddings ), "Padding_idx must be within num_embeddings" padding_idx = self.num_embeddings + padding_idx self.padding_idx = padding_idx self.init_method = init_method self.amp_enabled = amp_enabled assert num_embeddings > 0 self.weight = nn.Parameter( flow.empty( (num_embeddings, embedding_dim), dtype=flow.float32, placement=dist.get_layer_placement(layer_idx), sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), ) ) self.init_method(self.weight) # FIXME(lxy): Fill padding_idx is not supported in nd_sbp right now. # self._fill_padding_idx_with_zero() def forward(self, input_ids): weight = flow._C.amp_white_identity(self.weight) if self.amp_enabled else self.weight # embeddings with sbp sign: [B, B] # [B, B] x [S(0), B] --> [S(0), B] # ↑ ↑ ↑ # embed pos_ids pos_embed input_embeds = flow._C.gather(weight, input_ids, axis=0) return input_embeds def _fill_padding_idx_with_zero(self) -> None: if self.padding_idx is not None: with flow.no_grad(): self.weight[self.padding_idx] = flow.zeros( self.embedding_dim, placement=dist.get_layer_placement(0), sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), ) def extra_repr(self) -> str: s = "num_embeddings={num_embeddings}, embedding_dim={embedding_dim}" if self.padding_idx is not None: s += ", padding_idx={padding_idx}" return s.format(**self.__dict__)
4,527
34.936508
95
py
libai
libai-main/projects/T5/models/logits.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow from oneflow import nn from libai.layers import Linear from libai.utils import distributed as dist class LMLogits(nn.Module): def __init__(self, vocab_size, hidden_size=None, bias=False, model_type="t5", layer_idx=-1): super().__init__() self.model_type = model_type if model_type == "t5": self.bias = ( nn.Parameter( flow.zeros( (vocab_size,), dtype=flow.float32, placement=dist.get_layer_placement(layer_idx), sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.split(0)]), ) ) if bias else None ) elif model_type == "mt5": self.linear = Linear(hidden_size, vocab_size, bias=False, layer_idx=layer_idx) def forward(self, input, word_embeddings=None): if self.model_type == "t5": w = word_embeddings.to_global(placement=input.placement) input = input.to_global(grad_sbp=input.sbp) logits = flow._C.matmul(input, w, transpose_b=True) if self.bias is not None: logits = logits + self.bias else: logits = self.linear(input) return logits
1,958
35.962264
96
py
libai
libai-main/projects/T5/models/t5_model.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow import oneflow.nn as nn from libai.config import configurable from libai.layers import Linear, LMLogits from libai.models.t5_model import T5Loss from libai.models.utils import init_method_normal, scaled_init_method_normal from libai.utils import distributed as dist from projects.MT5.utils.mt5_loader import T5LoaderHuggerFace from projects.T5.models.embedding import T5Embedding from projects.T5.models.layer_norm import LayerNorm from projects.T5.models.transformer_layer import TransformerLayer from projects.T5.utils.mask import ExtendedMask class T5Model(flow.nn.Module): @configurable def __init__( self, vocab_size, hidden_size, hidden_layers, num_attention_heads, head_size, intermediate_size, embedding_dropout_prob, hidden_dropout_prob, attention_probs_dropout_prob, relative_attention_num_buckets, initializer_range=0.02, layernorm_eps=1e-12, amp_enabled=False, model_type="t5", ) -> None: super().__init__() self.model_type = model_type init_method = init_method_normal(initializer_range) scaled_init_method = scaled_init_method_normal(initializer_range, hidden_layers) self.embedding = T5Embedding( hidden_size=hidden_size, vocab_size=vocab_size, embedding_dropout_prob=embedding_dropout_prob, init_method=init_method, amp_enabled=amp_enabled, ) self.extended_attn_mask = ExtendedMask() encoder_layers = flow.nn.ModuleList( [ TransformerLayer( hidden_size=hidden_size, ffn_hidden_size=intermediate_size, num_attention_heads=num_attention_heads, head_size=head_size, relative_attention_num_buckets=relative_attention_num_buckets, is_decoder=False, attention_dropout_prob=attention_probs_dropout_prob, output_dropout_prob=hidden_dropout_prob, layernorm_epsilon=layernorm_eps, init_method=init_method, output_layer_init_method=scaled_init_method, layer_idx=i, model_type=model_type, has_relative_attention_bias=bool(i == 0), ) for i in range(hidden_layers) ] ) encoder_final_layernorm = LayerNorm( (hidden_size,), eps=layernorm_eps, layer_idx=hidden_layers - 1, ) self.encoder = flow.nn.Sequential() self.encoder.add_module("layers", encoder_layers) self.encoder.add_module("final_layernorm", encoder_final_layernorm) decoder_layers = flow.nn.ModuleList( [ TransformerLayer( hidden_size=hidden_size, ffn_hidden_size=intermediate_size, num_attention_heads=num_attention_heads, head_size=head_size, relative_attention_num_buckets=relative_attention_num_buckets, is_decoder=True, attention_dropout_prob=attention_probs_dropout_prob, output_dropout_prob=hidden_dropout_prob, layernorm_epsilon=layernorm_eps, init_method=init_method, output_layer_init_method=scaled_init_method, layer_idx=i, model_type=model_type, has_relative_attention_bias=bool(i - hidden_layers == 0), ) for i in range(hidden_layers, 2 * hidden_layers) ] ) decoder_final_layernorm = LayerNorm( (hidden_size,), eps=layernorm_eps, layer_idx=2 * hidden_layers - 1, ) self.decoder = flow.nn.Sequential() self.decoder.add_module("layers", decoder_layers) self.decoder.add_module("final_layernorm", decoder_final_layernorm) self.past_key_values = [None] * len(self.decoder.layers) self.encoder_states = None self.past_length = 0 if model_type == "mt5": self.lm_head = Linear( hidden_size, vocab_size, bias=False, layer_idx=2 * hidden_layers - 1 ) else: self.lm_head = LMLogits(vocab_size, bias=False) @classmethod def from_config(cls, cfg): return { "vocab_size": cfg.vocab_size, "hidden_size": cfg.hidden_size, "hidden_layers": cfg.hidden_layers, "num_attention_heads": cfg.num_attention_heads, "head_size": cfg.head_size, "intermediate_size": cfg.intermediate_size, "embedding_dropout_prob": cfg.embedding_dropout_prob, "hidden_dropout_prob": cfg.hidden_dropout_prob, "attention_probs_dropout_prob": cfg.attention_probs_dropout_prob, "relative_attention_num_buckets": cfg.relative_attention_num_buckets, "initializer_range": cfg.initializer_range, "layernorm_eps": cfg.layernorm_eps, "amp_enabled": cfg.amp_enabled, "model_type": cfg.model_type, } def forward( self, encoder_input_ids, decoder_input_ids, encoder_attn_mask, decoder_attn_mask, encoder_decoder_attn_mask, use_cache=False, ): encoder_input_ids = encoder_input_ids.to_global(placement=dist.get_layer_placement(0)) decoder_input_ids = decoder_input_ids.to_global(placement=dist.get_layer_placement(0)) encoder_attn_mask = encoder_attn_mask.to_global(placement=dist.get_layer_placement(0)) decoder_attn_mask = decoder_attn_mask.to_global(placement=dist.get_layer_placement(0)) encoder_decoder_attn_mask = encoder_decoder_attn_mask.to_global( placement=dist.get_layer_placement(0) ) if use_cache and self.encoder_states is not None: encoder_states = self.encoder_states else: position_bias = None encoder_decoder_position_bias = None self.set_cache(encoder_states=None, past_key_values=None) encoder_attn_mask = self.extended_attn_mask(encoder_attn_mask) enc_embedding_output = self.embedding(encoder_input_ids) enc_hidden_states = enc_embedding_output for layer in self.encoder.layers: enc_hidden_states, position_bias = layer( enc_hidden_states, encoder_attn_mask, position_bias=position_bias, ) encoder_states = self.encoder.final_layernorm(enc_hidden_states) decoder_attn_mask = self.extended_attn_mask( decoder_attn_mask, decoder_input_ids, is_decoder=True ) encoder_decoder_attn_mask = self.extended_attn_mask(encoder_decoder_attn_mask) dec_embedding_output = self.embedding(decoder_input_ids) dec_hidden_states = dec_embedding_output if use_cache: presents = [] position_bias = None encoder_decoder_position_bias = None for layer, past_key_value in zip(self.decoder.layers, self.past_key_values): dec_hidden_states, position_bias, encoder_decoder_position_bias = layer( dec_hidden_states, decoder_attn_mask, encoder_states, encoder_decoder_attn_mask, past_key_value=past_key_value, position_bias=position_bias, encoder_decoder_position_bias=encoder_decoder_position_bias, use_cache=use_cache, ) if use_cache: dec_hidden_states, present = dec_hidden_states presents.append(present) if use_cache: self.set_cache(encoder_states, past_key_values=presents) decoder_states = self.decoder.final_layernorm(dec_hidden_states) if self.model_type == "mt5": logits = self.lm_head(decoder_states) else: logits = self.lm_head(decoder_states, self.embedding.word_embeddings.weight) return logits def set_cache(self, encoder_states, past_key_values): self.encoder_states = encoder_states self.past_length = 0 if past_key_values is None else past_key_values[0][0].shape[2] if past_key_values is None: past_key_values = [None] * len(self.decoder.layers) assert len(past_key_values) == len(self.decoder.layers), ( f"past_key_values's length {len(past_key_values)} doesn't match " f"decoder num_layers' length {self.decoder.layers}" ) self.past_key_values = past_key_values class T5ForPreTraining(flow.nn.Module): def __init__(self, cfg) -> None: super().__init__() if cfg.pretrained_model_path is not None: loader = T5LoaderHuggerFace(T5Model, cfg, cfg.pretrained_model_path) self.t5_model = loader.load() else: self.t5_model = T5Model(cfg) self.loss_func = T5Loss() def set_cache(self, encoder_states, past_key_values): self.t5_model.set_cache(encoder_states, past_key_values) def forward( self, encoder_input_ids, decoder_input_ids, encoder_attn_mask, decoder_attn_mask, encoder_decoder_attn_mask, lm_labels=None, loss_mask=None, use_cache=False, ): logits = self.t5_model( encoder_input_ids, decoder_input_ids, encoder_attn_mask, decoder_attn_mask, encoder_decoder_attn_mask, use_cache=use_cache, ) if lm_labels is not None: lm_loss = self.loss_func(logits, lm_labels, loss_mask) return lm_loss else: return { "prediction_scores": logits, } @staticmethod def set_pipeline_stage_id(model): dist_utils = dist.get_dist_util() # Set pipeline parallelism stage_id if hasattr(model.t5_model.encoder.final_layernorm, "config"): # Old API in OneFlow 0.8 for module_block in model.modules(): if isinstance(module_block.origin, T5Embedding): module_block.config.set_stage( dist_utils.get_layer_stage_id(0), dist.get_layer_placement(0) ) elif isinstance(module_block.origin, ExtendedMask): module_block.config.set_stage( dist_utils.get_layer_stage_id(0), dist.get_layer_placement(0) ) elif isinstance(module_block.origin, TransformerLayer): module_block.config.set_stage( dist_utils.get_layer_stage_id(module_block.layer_idx), dist.get_layer_placement(module_block.layer_idx), ) elif isinstance(module_block.origin, T5Loss): module_block.config.set_stage( dist_utils.get_layer_stage_id(-1), dist.get_layer_placement(-1) ) model.t5_model.encoder.final_layernorm.config.set_stage( dist_utils.get_layer_stage_id(model.t5_model.encoder.final_layernorm.layer_idx), dist.get_layer_placement(model.t5_model.encoder.final_layernorm.layer_idx), ) model.t5_model.decoder.final_layernorm.config.set_stage( dist_utils.get_layer_stage_id(model.t5_model.decoder.final_layernorm.layer_idx), dist.get_layer_placement(model.t5_model.decoder.final_layernorm.layer_idx), ) else: for module_block in model.modules(): if isinstance(module_block.to(nn.Module), T5Embedding): module_block.to(flow.nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(0), dist.get_layer_placement(0) ) elif isinstance(module_block.to(nn.Module), ExtendedMask): module_block.to(flow.nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(0), dist.get_layer_placement(0) ) elif isinstance(module_block.to(nn.Module), TransformerLayer): module_block.to(flow.nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(module_block.layer_idx), dist.get_layer_placement(module_block.layer_idx), ) elif isinstance(module_block.to(nn.Module), T5Loss): module_block.to(flow.nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(-1), dist.get_layer_placement(-1) ) model.t5_model.encoder.final_layernorm.to(flow.nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(model.t5_model.encoder.final_layernorm.layer_idx), dist.get_layer_placement(model.t5_model.encoder.final_layernorm.layer_idx), ) model.t5_model.decoder.final_layernorm.to(flow.nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(model.t5_model.decoder.final_layernorm.layer_idx), dist.get_layer_placement(model.t5_model.decoder.final_layernorm.layer_idx), )
14,346
40.585507
96
py
libai
libai-main/projects/T5/models/mlp.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from oneflow import nn from libai.layers import Linear, build_activation class T5MLP(nn.Module): def __init__( self, hidden_size, ffn_hidden_size, output_dropout_prob=0.0, init_method=nn.init.xavier_normal_, output_layer_init_method=None, *, layer_idx=0, ): super().__init__() self.output_dropout_prob = output_dropout_prob if output_layer_init_method is None: output_layer_init_method = init_method self.dense_h_to_4h = Linear( hidden_size, ffn_hidden_size, bias=False, parallel="col", skip_bias_add=False, init_method=init_method, layer_idx=layer_idx, ) self.activation_func = build_activation("relu") self.dense_4h_to_h = Linear( ffn_hidden_size, hidden_size, bias=False, parallel="row", skip_bias_add=False, init_method=output_layer_init_method, layer_idx=layer_idx, ) self.dropout = nn.Dropout(self.output_dropout_prob) def forward(self, hidden_states): intermediate = self.dense_h_to_4h(hidden_states) intermediate = self.activation_func(intermediate) output = self.dense_4h_to_h(intermediate) output = self.dropout(output) return output class MT5MLP(nn.Module): def __init__( self, hidden_size, ffn_hidden_size, output_dropout_prob=0.0, init_method=nn.init.xavier_normal_, output_layer_init_method=None, *, layer_idx=0, ): super().__init__() self.output_dropout_prob = output_dropout_prob if output_layer_init_method is None: output_layer_init_method = init_method self.wi_0 = Linear( hidden_size, ffn_hidden_size, bias=False, parallel="col", skip_bias_add=False, init_method=init_method, layer_idx=layer_idx, ) self.wi_1 = Linear( hidden_size, ffn_hidden_size, bias=False, parallel="col", skip_bias_add=False, init_method=init_method, layer_idx=layer_idx, ) self.activation_func = build_activation("gelu_tanh") self.wo = Linear( ffn_hidden_size, hidden_size, bias=False, parallel="row", skip_bias_add=False, init_method=output_layer_init_method, layer_idx=layer_idx, ) self.dropout = nn.Dropout(self.output_dropout_prob) def forward(self, hidden_states): wi_0_out = self.wi_0(hidden_states) hidden_gelu = self.activation_func(wi_0_out) hidden_linear = self.wi_1(hidden_states) hidden_states = hidden_gelu * hidden_linear output = self.wo(hidden_states) output = self.dropout(output) return output
3,696
27.658915
74
py
libai
libai-main/projects/T5/models/transformer_layer.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow.nn as nn from libai.layers.droppath import DropPath from libai.utils import distributed as dist from projects.T5.models.attention import MultiheadAttention from projects.T5.models.layer_norm import LayerNorm from projects.T5.models.mlp import MT5MLP, T5MLP class TransformerLayer(nn.Module): """A single transformer layer. Transformer layer takes input with size [bsz, seq_length, hidden size] and returns an output of the same size. The input and output has same sbp sign, (S(0), B). Arguments: hidden_size: size of hidden state. ffn_hidden_size: size of feed forword neural network. num_attention_heads: number of attention heads. is_decoder: used to specify whether this is transformer encoder layer or transformer decoder layer. Default: ``False``. attention_dropout_prob: dropout probability of attention weights. output_dropout_prob: dropout probability of output. layernorm_epsilon: epsilon used in layernorm layer. Default: `1e-5`. init_method: method to initialize the input layer weights. output_layer_init_method: method to initialize the output layer weights. If None, use `init_method`. layer_idx: the layer index, which determines the placement. """ def __init__( self, hidden_size, ffn_hidden_size, num_attention_heads, head_size, relative_attention_num_buckets, is_decoder=False, attention_dropout_prob=0.0, output_dropout_prob=0.0, drop_path_prob=0.0, layernorm_epsilon=1e-5, init_method=nn.init.xavier_normal_, output_layer_init_method=None, *, layer_idx=0, model_type="t5", has_relative_attention_bias=False ): super().__init__() self.hidden_size = hidden_size self.ffn_hidden_size = ffn_hidden_size self.num_attention_heads = num_attention_heads self.head_size = head_size self.attention_dropout_prob = attention_dropout_prob self.output_dropout_prob = output_dropout_prob self.layernorm_epsilon = layernorm_epsilon self.layer_idx = layer_idx self.is_decoder = is_decoder self.init_method = init_method if output_layer_init_method is None: output_layer_init_method = init_method self.output_layer_init_method = output_layer_init_method self.drop_path = DropPath(drop_path_prob) if drop_path_prob > 0.0 else nn.Identity() self.input_layernorm = LayerNorm( self.hidden_size, eps=self.layernorm_epsilon, layer_idx=self.layer_idx ) self.self_attention = self.build_attention( is_cross_attention=False, relative_attention_num_buckets=relative_attention_num_buckets, has_relative_attention_bias=has_relative_attention_bias, is_decoder=self.is_decoder, ) self.post_attention_layernorm = LayerNorm( self.hidden_size, eps=self.layernorm_epsilon, layer_idx=self.layer_idx ) if self.is_decoder: self.cross_attention = self.build_attention( is_cross_attention=True, relative_attention_num_buckets=relative_attention_num_buckets, is_decoder=self.is_decoder, ) self.post_cross_attention_layernorm = LayerNorm( self.hidden_size, eps=self.layernorm_epsilon, layer_idx=self.layer_idx ) if model_type == "mt5": self.mlp = MT5MLP( self.hidden_size, self.ffn_hidden_size, self.output_dropout_prob, self.init_method, output_layer_init_method=self.output_layer_init_method, layer_idx=self.layer_idx, ) elif model_type == "t5": self.mlp = T5MLP( self.hidden_size, self.ffn_hidden_size, self.output_dropout_prob, self.init_method, output_layer_init_method=self.output_layer_init_method, layer_idx=self.layer_idx, ) def forward( self, hidden_states, attention_mask=None, encoder_states=None, encoder_attention_mask=None, past_key_value=None, use_cache=False, position_bias=None, encoder_decoder_position_bias=None, ): """ Args: hidden_states: shape is (batch_size, seq_length, hidden_size), sbp signature is (S(0), B). attention_mask: the combination of key padding mask and casual mask of hidden states with shape (batch_size, 1, seq_length, seq_length) and the sbp signature is (S(0), B), encoder_states: encoder output with shape (batch_size, seq_length, hidden_size) and the sbp signature is (S(0), B), which will be used in cross attention. encoder_attention_mask: key padding mask of encoder states with shape (batch_size, 1, seq_length, seq_length) and the sbp signature is (S(0), B). past_key_value: tuple of key and value, each shape is (seq_length, bsz, num_heads, head_size), For decoder layer, the past_key_value contains the states both from self attention and cross attention. use_cache: it will be set to `True` when the model is in the inference phase and used for incremental decoding. """ # Change placement for pipeline parallelsim hidden_states = hidden_states.to_global(placement=dist.get_layer_placement(self.layer_idx)) # hidden_states shape: (batch_size, seq_length, hidden_size) if attention_mask is not None: attention_mask = attention_mask.to_global( placement=dist.get_layer_placement(self.layer_idx) ) if past_key_value is not None: if self.is_decoder: assert len(past_key_value) == 4 self_attn_past_key_value = past_key_value[:2] cross_attn_past_key_value = past_key_value[2:] else: self_attn_past_key_value = past_key_value cross_attn_past_key_value = None else: self_attn_past_key_value, cross_attn_past_key_value = None, None layernorm_output = self.input_layernorm(hidden_states) attention_output, position_bias = self.self_attention( layernorm_output, attention_mask=attention_mask, past_key_value=self_attn_past_key_value, position_bias=position_bias, use_cache=use_cache, ) attention_output = self.drop_path(attention_output) if use_cache: attention_output, presents = attention_output else: presents = None hidden_states = hidden_states + attention_output layernorm_output = self.post_attention_layernorm(hidden_states) if self.is_decoder: if presents is not None: query_length = presents[0].shape[2] else: query_length = None attention_output, encoder_decoder_position_bias = self.cross_attention( layernorm_output, encoder_states, attention_mask=encoder_attention_mask, past_key_value=cross_attn_past_key_value, position_bias=encoder_decoder_position_bias, use_cache=use_cache, query_length=query_length, ) if use_cache: attention_output, decoder_presents = attention_output presents = presents + decoder_presents attention_output = self.drop_path(attention_output) hidden_states = hidden_states + attention_output layernorm_output = self.post_cross_attention_layernorm(hidden_states) mlp_output = self.mlp(layernorm_output) mlp_output = self.drop_path(mlp_output) output = hidden_states + mlp_output if use_cache: output = (output, presents) output = (output,) + (position_bias,) if self.is_decoder: output = output + (encoder_decoder_position_bias,) return output def build_attention( self, is_cross_attention=False, relative_attention_num_buckets=None, has_relative_attention_bias=False, is_decoder=False, ): return MultiheadAttention( self.hidden_size, self.num_attention_heads, head_size=self.head_size, relative_attention_num_buckets=relative_attention_num_buckets, is_cross_attention=is_cross_attention, attention_dropout_prob=self.attention_dropout_prob, output_dropout_prob=self.output_dropout_prob, init_method=self.init_method, output_layer_init_method=self.output_layer_init_method, layer_idx=self.layer_idx, has_relative_attention_bias=has_relative_attention_bias, is_decoder=is_decoder, )
9,934
38.268775
99
py
libai
libai-main/projects/T5/models/layer_norm.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow from libai.utils import distributed as dist class LayerNorm(flow.nn.Module): def __init__(self, normalized_shape, eps=1e-6, layer_idx=0): super().__init__() self.layer_idx = layer_idx self.weight = flow.nn.Parameter( flow.ones( normalized_shape, dtype=flow.float32, placement=dist.get_layer_placement(layer_idx), sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), ) ) self.l2norm_epsilon = eps def forward(self, hidden_states): return flow._C.rms_norm(hidden_states, self.weight, self.weight.shape, self.l2norm_epsilon)
1,324
34.810811
99
py
libai
libai-main/projects/T5/models/attention.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import math from typing import Tuple import oneflow as flow from oneflow import nn from libai.layers.linear import Linear from libai.utils import distributed as dist from projects.T5.models.embedding import Embedding class MultiheadAttention(nn.Module): """Multi-head attention layer, support self attention and cross attention. Args: hidden_size: size of hidden state. num_attention_heads: number of attention heads. is_cross_attention: used to specify whether it is self attention or cross attention. Defaults to False. attention_dropout_prob: dropout probability of attention weights. Defaults to 0.0. output_dropout_prob: dropout probability of output. Defaults to 0.0. init_method: method to initialize the input layer weights. Defaults to ``init.xavier_normal_``. output_layer_init_method: method to initialize the output layer weights. If None, use ``init_method``. layer_idx: a layer_idx sign which determines the placements. It will be used in pipeline parallelism. Defaults to 0. """ def __init__( self, hidden_size, num_attention_heads, head_size, relative_attention_num_buckets, is_cross_attention=False, attention_dropout_prob=0.0, output_dropout_prob=0.0, init_method=nn.init.xavier_normal_, output_layer_init_method=None, *, layer_idx=0, has_relative_attention_bias=False, is_decoder=False, ): super().__init__() self.hidden_size = hidden_size self.relative_attention_num_buckets = relative_attention_num_buckets self.has_relative_attention_bias = has_relative_attention_bias self.is_decoder = is_decoder self.attention_dropout_prob = attention_dropout_prob if output_layer_init_method is None: output_layer_init_method = init_method self.num_heads = num_attention_heads self.head_size = head_size self.dropout = nn.Dropout(p=attention_dropout_prob) self.norm_factor = 1.0 / math.sqrt(float(self.head_size)) self.is_cross_attention = is_cross_attention self.output_dropout = nn.Dropout(p=output_dropout_prob) if self.is_cross_attention: self.query = Linear( self.hidden_size, self.num_heads * self.head_size, bias=False, parallel="col", init_method=init_method, layer_idx=layer_idx, ) self.key_value = Linear( self.hidden_size, self.num_heads * self.head_size * 2, bias=False, parallel="col", init_method=init_method, layer_idx=layer_idx, ) else: self.query_key_value = Linear( self.hidden_size, self.num_heads * self.head_size * 3, bias=False, parallel="col", init_method=init_method, layer_idx=layer_idx, ) self.dense = Linear( self.num_heads * self.head_size, self.hidden_size, bias=False, parallel="row", init_method=output_layer_init_method, skip_bias_add=False, layer_idx=layer_idx, ) if self.has_relative_attention_bias: self.relative_attention_bias = Embedding( self.relative_attention_num_buckets, self.num_heads, layer_idx=layer_idx ) def forward( self, hidden_states: flow.Tensor, encoder_states: flow.Tensor = None, attention_mask: flow.Tensor = None, past_key_value: Tuple[flow.Tensor, flow.Tensor] = None, use_cache: bool = False, position_bias=None, query_length=None, ): """ Args: hidden_states (flow.Tensor): shape is [bsz, tgt_len, hidden_size]. encoder_states (flow.Tensor, optional): shape is [bsz, src_len, hidden_size]. Defaults to None. attention_mask (flow.Tensor, optional): shape is [bsz, 1, tgt_len, src_len]. It should be the combination of padding mask and casual mask. It is the padding mask of source input when used with self-attention in encoder. And it is the combination of padding mask of target input and casual mask when used with self-attention in decoder. It is the padding mask of source input when used with cross-attention in decoder. Defaults to None. past_key_value (Tuple[flow.Tensor, flow.Tensor], optional): tuple of key and value, each shape is [bsz, num_heads, src_len, head_size]. Defaults to None. use_cache (bool, optional): it will be set to True, when the model is in the inference phase and used for incremental decoding. Defaults to False. """ # hidden_states, encoder_states: [S(0), B] # attention_mask: [S(0), B] if encoder_states is not None: encoder_states = encoder_states.to_global(placement=hidden_states.placement) if attention_mask is not None: attention_mask = attention_mask.to_global(placement=hidden_states.placement) bsz, real_seq_length = hidden_states.size()[:2] if past_key_value is not None: assert ( len(past_key_value) == 2 ), "past_key_value should have 2 past states: keys and values." f"Got {len(past_key_value)} past states.\n" real_seq_length += past_key_value[0].shape[2] if query_length is None else query_length key_length = real_seq_length if encoder_states is None else encoder_states.shape[1] if self.is_cross_attention: # if it is cross attention, key and value should be calculated only once, and the # result can be reused. query = self.query(hidden_states) query = query.view(bsz, -1, self.num_heads, self.head_size) query = query.permute(0, 2, 1, 3) if past_key_value is not None: key, value = past_key_value elif encoder_states is not None: key_value = self.key_value(encoder_states) key_value = key_value.view(bsz, -1, self.num_heads, 2 * self.head_size) key_value = key_value.permute(0, 2, 1, 3) key, value = flow.chunk(key_value, chunks=2, dim=-1) else: raise ValueError( "past_key_value and encoder_states cannot be None at the same time." ) else: # if it is self attention, query, key, and value are all obtained from hidden_states. # when in the inference phase of an incremental decoder, # hidden_states is the last-added state, # the full key and value could be obtained by concatenating with past_key_value. query_key_value = self.query_key_value(hidden_states) query_key_value = query_key_value.view(bsz, -1, self.num_heads, 3 * self.head_size) query_key_value = query_key_value.permute( 0, 2, 1, 3 ) # [bsz, num_heads, src_len, 3 * head_size] query, key, value = flow.chunk(query_key_value, chunks=3, dim=-1) if past_key_value is not None: past_key, past_value = past_key_value key = flow.cat((past_key.type_as(key), key), dim=2) value = flow.cat((past_value.type_as(value), value), dim=2) # query, key, value: [S(0), S(1)], shape: [bsz, num_heads, seq_length, head_size] if use_cache: past_key_value = (key, value) # [bsz, num_heads, tgt_len, src_len] with [S(0), S(1)] attention_scores = flow.matmul(query, key, transpose_b=True) if position_bias is None: if not self.has_relative_attention_bias: position_bias = flow.zeros( (1, self.num_heads, real_seq_length, key_length), sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=attention_scores.placement, ) else: position_bias = self.compute_bias( real_seq_length, key_length, placement=attention_mask.placement ) if past_key_value is not None: position_bias = position_bias[:, :, -hidden_states.size(1) :, :] position_bias = position_bias + (1 - attention_mask) * -1000 position_bias = position_bias.to_global(placement=attention_scores.placement) attention_scores = attention_scores + position_bias # [S(0), S(1)] x [S(0), B] = [S(0), S(1)] if attention_mask is not None: attention_scores = flow.mul(attention_scores, attention_mask) attention_scores = attention_scores - 10000.0 * (1 - attention_mask) # TODO(xingyu.liao): graph will occur `where_scalar` errors # when using `masked_fill` # attention_scores = attention_scores.masked_fill(1 - attention_mask, -10000.0) attention_weights = flow.softmax(attention_scores, dim=-1) # [bsz, num_heads, tgt_len, src_len] attention_weights = self.dropout(attention_weights) else: attention_weights = flow.softmax(attention_scores, dim=-1) # [bsz, num_heads, tgt_len, src_len] attention_weights = self.dropout(attention_weights) # Context shape: [bsz, num_heads, tgt_len, head_size] with [S(0), S(1)] context = flow.matmul(attention_weights, value) # Change shape: [bsz, num_heads, tgt_len, head_size] -> [bsz, tgt_len, num_heads, head_size] context = context.transpose(1, 2) # Concat multi-head results from # [bsz, tgt_len, num_heads, head_size] -> [bsz, tgt_len, num_heads * head_size] # SBP sign: [S(0), S(2)] # [S(0), S(2)] x [B, S(0)] = [S(0), P] -> [S(0), B] output = self.dense(context.flatten(2)) output = self.output_dropout(output) if use_cache: output = (output, past_key_value) output = (output,) + (position_bias,) return output def extra_repr(self) -> str: return "hidden_size={}, num_heads={}, is_cross_attention={}".format( self.hidden_size, self.num_heads, self.is_cross_attention, ) def _relative_position_bucket( self, relative_position, bidirectional=True, num_buckets=32, max_distance=128 ): # relative_position: (seq_len, seq_len) relative_buckets = 0 if bidirectional: num_buckets //= 2 relative_buckets = ( relative_buckets + (relative_position > 0).to(flow.long) * num_buckets ) relative_position = flow.abs(relative_position) else: relative_position = ( -1 * flow.min( relative_position, flow.zeros( relative_position.size(), sbp=relative_position.sbp, placement=relative_position.placement, ), ).to(flow.long) ) max_exact = num_buckets // 2 is_small = relative_position < max_exact relative_postion_if_large = max_exact + ( flow.log(relative_position.float() / max_exact) / math.log(max_distance / max_exact) * (num_buckets - max_exact) ).to(flow.long) relative_postion_if_large = flow.min( relative_postion_if_large, flow.zeros( relative_postion_if_large.size(), dtype=relative_postion_if_large.dtype, sbp=relative_postion_if_large.sbp, placement=relative_postion_if_large.placement, ).fill_(num_buckets - 1), ) relative_buckets = relative_buckets + flow.where( is_small, relative_position, relative_postion_if_large ) return relative_buckets def compute_bias(self, query_length, key_length, placement=None): """Compute binned relative position bias""" context_position = flow.arange( query_length, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=placement, ) memory_position = flow.arange( key_length, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=placement, ) relative_position = ( memory_position[None, :] - context_position[:, None] ) # shape (query_length, key_length) relative_position_bucket = self._relative_position_bucket( relative_position, bidirectional=(not self.is_decoder), num_buckets=self.relative_attention_num_buckets, ) # shape (query_length, key_length) values = self.relative_attention_bias( relative_position_bucket ) # shape (query_length, key_length, num_heads) values = values.permute([2, 0, 1]).unsqueeze( 0 ) # shape (1, num_heads, query_length, key_length) return values
14,317
39.792023
100
py
libai
libai-main/projects/T5/datasets/dataset.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import warnings try: import datasets except: # noqa warnings.warn("datasets library is needed") import numpy as np import oneflow as flow from oneflow.utils.data import Dataset from libai.data.structures import DistTensorData, Instance def get_data(path): total_data = [] for i in range(10): path = path[:-1] + str(i) dataset = datasets.load_from_disk(path) for i in dataset: total_data.append(i) return total_data def compute_input_and_target_lengths(inputs_length, noise_density, mean_noise_span_length): """This function is copy of `random_spans_helper <https://github.com/google-research/ text-to-text-transfer-transformer/blob/84f8bcc14b5f2c03de51bd3587609ba8f6bbd1cd/ t5/data/preprocessors.py#L2466>`__ . Training parameters to avoid padding with random_spans_noise_mask. When training a model with random_spans_noise_mask, we would like to set the other training hyperparmeters in a way that avoids padding. This function helps us compute these hyperparameters. We assume that each noise span in the input is replaced by extra_tokens_per_span_inputs sentinel tokens, and each non-noise span in the targets is replaced by extra_tokens_per_span_targets sentinel tokens. This function tells us the required number of tokens in the raw example (for split_tokens()) as well as the length of the encoded targets. Note that this function assumes the inputs and targets will have EOS appended and includes that in the reported length. Args: inputs_length: an integer - desired length of the tokenized inputs sequence noise_density: a float mean_noise_span_length: a float Returns: tokens_length: length of original text in tokens targets_length: an integer - length in tokens of encoded targets sequence """ def _tokens_length_to_inputs_length_targets_length(tokens_length): num_noise_tokens = int(round(tokens_length * noise_density)) num_nonnoise_tokens = tokens_length - num_noise_tokens num_noise_spans = int(round(num_noise_tokens / mean_noise_span_length)) # inputs contain all nonnoise tokens, sentinels for all noise spans # and one EOS token. _input_length = num_nonnoise_tokens + num_noise_spans + 1 _output_length = num_noise_tokens + num_noise_spans + 1 return _input_length, _output_length tokens_length = inputs_length while _tokens_length_to_inputs_length_targets_length(tokens_length + 1)[0] <= inputs_length: tokens_length += 1 inputs_length, targets_length = _tokens_length_to_inputs_length_targets_length(tokens_length) # minor hack to get the targets length to be equal to inputs length # which is more likely to have been set to a nice round number. if noise_density == 0.5 and targets_length > inputs_length: tokens_length -= 1 targets_length -= 1 return tokens_length, targets_length class UnsuperviseT5Dataset(Dataset): """This function is copy of https://github.com/IDEA-CCNL/Fengshenbang-LM/blob/ ec13aeb8689cfafaa6a7a9e9595d110edbe34123/fengshen/data/t5_dataloader/t5_datasets.py#L61. """ def __init__(self, data_path): # [{input_ids: ...}, {input_ids: ...}, ...] self.data = get_data(data_path) def __len__(self): return len(self.data) def __getitem__(self, index): x = self.data[index] return x class collate_fn: def __init__( self, vocab_size, max_seq_length, noise_density, mean_noise_span_length, eos_token_id=1, pad_token_id=0, decoder_start_token_id=0, ): self.max_seq_length = max_seq_length self.eos_token_id = eos_token_id self.pad_token_id = pad_token_id self.decoder_start_token_id = decoder_start_token_id self.vocab_size = vocab_size self.noise_density = noise_density self.mean_noise_span_length = mean_noise_span_length self.expanded_inputs_length, self.targets_length = compute_input_and_target_lengths( inputs_length=self.max_seq_length, noise_density=self.noise_density, mean_noise_span_length=self.mean_noise_span_length, ) def __call__(self, examples): batch = { k: np.array([examples[i][k] for i in range(len(examples))]) for k, v in examples[0].items() } input_ids = np.array(batch["input_ids"]) batch_size, expanded_input_length = input_ids.shape mask_indices = np.asarray( [self.random_spans_noise_mask(expanded_input_length) for i in range(batch_size)] ) labels_mask = ~mask_indices input_ids_sentinel = self.create_sentinel_ids(mask_indices.astype(np.int8)) labels_sentinel = self.create_sentinel_ids(labels_mask.astype(np.int8)) batch["input_ids"] = self.filter_input_ids(input_ids, input_ids_sentinel) batch["labels"] = self.filter_input_ids(input_ids, labels_sentinel) if batch["input_ids"].shape[-1] != self.max_seq_length: raise ValueError( f"`input_ids` are incorrectly preprocessed. `input_ids` length is \ {batch['input_ids'].shape[-1]}, but should be {self.targets_length}." ) if batch["labels"].shape[-1] != self.targets_length: raise ValueError( f"`labels` are incorrectly preprocessed. `labels` length is \ {batch['labels'].shape[-1]}, but should be {self.targets_length}." ) batch["decoder_input_ids"] = self.shift_tokens_right( batch["labels"], self.pad_token_id, self.decoder_start_token_id ) return Instance( encoder_input_ids=DistTensorData(flow.tensor(batch["input_ids"])), decoder_input_ids=DistTensorData(flow.tensor(batch["decoder_input_ids"])), encoder_attn_mask=DistTensorData( flow.ones(len(batch["input_ids"]), len(batch["input_ids"][0])).to(flow.bool) ), decoder_attn_mask=DistTensorData( flow.ones(len(batch["decoder_input_ids"]), len(batch["decoder_input_ids"][0])).to( flow.bool ) ), encoder_decoder_attn_mask=DistTensorData( flow.ones( len(batch["input_ids"]), len(batch["decoder_input_ids"][0]), len(batch["input_ids"][0]), ).to(flow.bool) ), lm_labels=DistTensorData(flow.tensor(batch["labels"])), loss_mask=DistTensorData(flow.tensor(batch["labels"])), ) def filter_input_ids(self, input_ids, sentinel_ids): batch_size = input_ids.shape[0] input_ids_full = np.where(sentinel_ids != 0, sentinel_ids, input_ids) input_ids = input_ids_full[input_ids_full >= 0].reshape((batch_size, -1)) input_ids = np.concatenate( [input_ids, np.full((batch_size, 1), self.eos_token_id, dtype=np.int32)], axis=-1 ) return input_ids def create_sentinel_ids(self, mask_indices): start_indices = mask_indices - np.roll(mask_indices, 1, axis=-1) * mask_indices start_indices[:, 0] = mask_indices[:, 0] sentinel_ids = np.where( start_indices != 0, np.cumsum(start_indices, axis=-1), start_indices ) sentinel_ids = np.where(sentinel_ids != 0, (self.vocab_size - sentinel_ids), 0) sentinel_ids -= mask_indices - start_indices return sentinel_ids def random_spans_noise_mask(self, length): orig_length = length num_noise_tokens = int(np.round(length * self.noise_density)) # avoid degeneracy by ensuring positive numbers of noise and nonnoise tokens. num_noise_tokens = min(max(num_noise_tokens, 1), length - 1) num_noise_spans = int(np.round(num_noise_tokens / self.mean_noise_span_length)) # avoid degeneracy by ensuring positive number of noise spans num_noise_spans = max(num_noise_spans, 1) num_nonnoise_tokens = length - num_noise_tokens # pick the lengths of the noise spans and the non-noise spans def _random_segmentation(num_items, num_segments): mask_indices = np.arange(num_items - 1) < (num_segments - 1) np.random.shuffle(mask_indices) first_in_segment = np.pad(mask_indices, [[1, 0]]) segment_id = np.cumsum(first_in_segment) # count length of sub segments assuming that list is sorted _, segment_length = np.unique(segment_id, return_counts=True) return segment_length noise_span_lengths = _random_segmentation(num_noise_tokens, num_noise_spans) nonnoise_span_lengths = _random_segmentation(num_nonnoise_tokens, num_noise_spans) interleaved_span_lengths = np.reshape( np.stack([nonnoise_span_lengths, noise_span_lengths], axis=1), [num_noise_spans * 2] ) span_starts = np.cumsum(interleaved_span_lengths)[:-1] span_start_indicator = np.zeros((length,), dtype=np.int8) span_start_indicator[span_starts] = True span_num = np.cumsum(span_start_indicator) is_noise = np.equal(span_num % 2, 1) return is_noise[:orig_length] def shift_tokens_right( self, input_ids: np.array, pad_token_id: int, decoder_start_token_id: int ) -> np.ndarray: """ Shift input ids one token to the right. """ shifted_input_ids = np.zeros_like(input_ids) shifted_input_ids[:, 1:] = input_ids[:, :-1] shifted_input_ids[:, 0] = decoder_start_token_id shifted_input_ids = np.where(shifted_input_ids == -100, pad_token_id, shifted_input_ids) return shifted_input_ids
10,558
40.735178
98
py
libai
libai-main/projects/T5/configs/optim.py
import oneflow as flow from libai.optim import get_default_optimizer_params from libai.config import LazyCall optim = LazyCall(flow.optim.AdamW)( params=LazyCall(get_default_optimizer_params)( # params.model is meant to be set to the model object, # before instantiating the optimizer. clip_grad_max_norm=1.0, clip_grad_norm_type=2.0, weight_decay_norm=0.0, weight_decay_bias=0.0, ), lr=1e-4, weight_decay=0.01, betas=(0.9, 0.999), eps=1e-8, do_bias_correction=True, )
547
25.095238
62
py
libai
libai-main/projects/T5/configs/t5_model_config.py
from omegaconf import DictConfig cfg = dict( vocab_size=30522, hidden_size=768, hidden_layers=6, num_attention_heads=12, head_size=64, intermediate_size=1536, hidden_dropout_prob=0.1, attention_probs_dropout_prob=0.1, relative_attention_num_buckets=32, embedding_dropout_prob=0.1, initializer_range=0.02, layernorm_eps=1e-5, amp_enabled=False, model_type="t5", ) cfg = DictConfig(cfg)
444
19.227273
38
py
libai
libai-main/projects/T5/configs/mt5_pretrain.py
from libai import evaluation from libai.data.build import build_nlp_train_loader from omegaconf import OmegaConf from libai.config import LazyCall from libai.evaluation import PPLEvaluator, evaluator from libai.scheduler import WarmupExponentialLR from configs.common.train import train from configs.common.models.graph import graph from projects.T5.configs.optim import optim from projects.T5.configs.t5_model_config import cfg from projects.T5.datasets.dataset import UnsuperviseT5Dataset, collate_fn from projects.T5.models.t5_model import T5ForPreTraining train_data_path = "projects/T5/data/training_data/part_0" pretrained_model_path = None micro_batch_size = 64 optim["lr"] = 1e-4 # dataloader dataloader = OmegaConf.create() dataloader.train = LazyCall(build_nlp_train_loader)( dataset=[ LazyCall(UnsuperviseT5Dataset)( data_path=train_data_path, ) ], collate_fn=collate_fn( vocab_size=12902, max_seq_length=512, noise_density=0.15, mean_noise_span_length=3, eos_token_id=12801, pad_token_id=0, decoder_start_token_id=12800, ), ) model = LazyCall(T5ForPreTraining)(cfg=cfg) # model config model.cfg.vocab_size = 12902 model.cfg.hidden_size = 512 model.cfg.hidden_layers = 8 model.cfg.num_attention_heads = 6 model.cfg.head_size = 64 model.cfg.intermediate_size = 1024 model.cfg.hidden_dropout_prob = 0.0 model.cfg.attention_probs_dropout_prob = 0.0 model.cfg.embedding_dropout_prob = 0.0 model.cfg.layernorm_eps = 1e-6 model.cfg.model_type = "mt5" model.cfg.pretrained_model_path = pretrained_model_path train.update( dict( output_dir="projects/T5/output/mt5_output", train_micro_batch_size=micro_batch_size, train_epoch=1, train_iter=24000, log_period=10, amp=dict(enabled=False), warmup_ratio=1 / 24, # checkpointer=dict(period=10, max_to_keep=20), dist=dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, pipeline_num_layers=2 * model.cfg.hidden_layers, ), scheduler=LazyCall(WarmupExponentialLR)( warmup_factor=0.001, gamma=1.0, warmup_method="linear", warmup_iter=0.0, ), evaluation=dict( evaluator=LazyCall(PPLEvaluator)(), enabled=True, eval_iter=1e5, eval_period=5000, ), ) ) train.zero_optimization.enabled = True train.zero_optimization.stage = 2
2,566
26.902174
73
py
libai
libai-main/projects/T5/utils/weight_convert.py
import argparse import oneflow as flow import torch from libai.config import LazyConfig def parse_args(): parser = argparse.ArgumentParser(description="MT5 Weight Convertor") parser.add_argument( "--oneflow_state_dict_path", type=str, help="The path of mt5's checkpoint in LiBai" ) parser.add_argument( "--config_path", type=str, default="projects/T5/configs/mt5_pretrain.py", help="The path of the training config", ) parser.add_argument("--save_path", type=str, default="projects/T5/pytorch_model.bin") return parser.parse_args() def fix_qkv_ordering(qkv, head_size, num_heads, hidden_size=None): hidden_size = (head_size * num_heads) if hidden_size is None else hidden_size num_of_qkv = qkv.shape[0] // (head_size * num_heads) qkv = qkv.view(-1) qkv = qkv.view(num_heads, num_of_qkv, head_size, hidden_size) qkv = qkv.permute(1, 0, 2, 3).contiguous() qkv = qkv.view(num_of_qkv * head_size * num_heads, hidden_size) return qkv def convert_tensor(tensor): return torch.tensor(tensor.detach().to_numpy(), dtype=torch.float32) def convert_state_dict(oneflow_state_dict_path, libai_cfg, prefix="t5_model."): oneflow_state_dict = flow.load(oneflow_state_dict_path) torch_state_dict = {} # Get configs num_heads = libai_cfg.get("num_attention_heads") hidden_size = libai_cfg.get("hidden_size") head_size = libai_cfg.get("head_size", None) if head_size is None: head_size = int(hidden_size / num_heads) layer_idx = 3 if len(prefix) > 1 else 2 enc_dec_idx = 1 if len(prefix) > 1 else 0 op_idx = 4 if len(prefix) > 1 else 3 # Convert T5's Embedding layers. x = convert_tensor(oneflow_state_dict.pop(prefix + "embedding.word_embeddings.weight")) new_key = "shared.weight" torch_state_dict[new_key] = x new_key = "encoder.embed_tokens.weight" torch_state_dict[new_key] = x new_key = "decoder.embed_tokens.weight" torch_state_dict[new_key] = x # Convert T5's final_layer_norm new_key = "encoder.final_layer_norm.weight" torch_state_dict[new_key] = convert_tensor( oneflow_state_dict.pop(prefix + "encoder.final_layernorm.weight") ) new_key = "decoder.final_layer_norm.weight" torch_state_dict[new_key] = convert_tensor( oneflow_state_dict.pop(prefix + "decoder.final_layernorm.weight") ) old_keys = list(oneflow_state_dict.keys()) # Convert T5's lm_head new_key = "lm_head.weight" if prefix + new_key in old_keys: torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(prefix + new_key)) for key in old_keys: keys = key.split(".") if op_idx > len(keys): continue layers = keys[layer_idx] enc_dec = keys[enc_dec_idx] op_name = keys[op_idx] if keys[op_idx + 1] == "relative_attention_bias": new_key = enc_dec + ".block.0.layer.0.SelfAttention.relative_attention_bias.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) # Convert T5's Encoder layers. if enc_dec == "encoder": if op_name == "self_attention": if keys[op_idx + 1] == "query_key_value": x = oneflow_state_dict.pop(key) x = fix_qkv_ordering(x, head_size, num_heads, hidden_size) q, k, v = flow.chunk(x, chunks=3, dim=0) new_key = "encoder.block." + layers + ".layer.0.SelfAttention.q.weight" torch_state_dict[new_key] = convert_tensor(q) new_key = new_key.replace(".q", ".k") torch_state_dict[new_key] = convert_tensor(k) new_key = new_key.replace(".k", ".v") torch_state_dict[new_key] = convert_tensor(v) if keys[op_idx + 1] == "dense": new_key = "encoder.block." + layers + ".layer.0.SelfAttention.o.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif op_name == "input_layernorm": new_key = "encoder.block." + layers + ".layer.0.layer_norm.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif op_name == "post_attention_layernorm": new_key = "encoder.block." + layers + ".layer.1.layer_norm.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif op_name == "mlp": if libai_cfg.get("model_type") == "mt5": if keys[op_idx + 1] == "wi_0": new_key = "encoder.block." + layers + ".layer.1.DenseReluDense.wi_0.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) if keys[op_idx + 1] == "wi_1": new_key = "encoder.block." + layers + ".layer.1.DenseReluDense.wi_1.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) if keys[op_idx + 1] == "wo": new_key = "encoder.block." + layers + ".layer.1.DenseReluDense.wo.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif libai_cfg.get("model_type") == "t5": if keys[op_idx + 1] == "dense_h_to_4h": new_key = "encoder.block." + layers + ".layer.1.DenseReluDense.wi.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) if keys[op_idx + 1] == "dense_4h_to_h": new_key = "encoder.block." + layers + ".layer.1.DenseReluDense.wo.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) # Convert T5's decoder Layers. elif enc_dec == "decoder": if op_name == "self_attention": if keys[op_idx + 1] == "query_key_value": x = oneflow_state_dict.pop(key) x = fix_qkv_ordering(x, head_size, num_heads, hidden_size) q, k, v = flow.chunk(x, chunks=3, dim=0) new_key = "decoder.block." + layers + ".layer.0.SelfAttention.q.weight" torch_state_dict[new_key] = convert_tensor(q) new_key = new_key.replace(".q", ".k") torch_state_dict[new_key] = convert_tensor(k) new_key = new_key.replace(".k", ".v") torch_state_dict[new_key] = convert_tensor(v) if keys[op_idx + 1] == "dense": new_key = "decoder.block." + layers + ".layer.0.SelfAttention.o.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif op_name == "input_layernorm": new_key = "decoder.block." + layers + ".layer.0.layer_norm.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif op_name == "post_attention_layernorm": new_key = "decoder.block." + layers + ".layer.1.layer_norm.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif op_name == "post_cross_attention_layernorm": new_key = "decoder.block." + layers + ".layer.2.layer_norm.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif op_name == "cross_attention": if keys[op_idx + 1] == "query": x = oneflow_state_dict.pop(key) x = fix_qkv_ordering(x, head_size, num_heads, hidden_size) new_key = "decoder.block." + layers + ".layer.1.EncDecAttention.q.weight" torch_state_dict[new_key] = convert_tensor(x) if keys[op_idx + 1] == "key_value": x = oneflow_state_dict.pop(key) x = fix_qkv_ordering(x, head_size, num_heads, hidden_size) k, v = flow.chunk(x, chunks=2, dim=0) new_key = "decoder.block." + layers + ".layer.1.EncDecAttention.k.weight" torch_state_dict[new_key] = convert_tensor(k) new_key = new_key.replace(".k", ".v") torch_state_dict[new_key] = convert_tensor(v) if keys[op_idx + 1] == "dense": new_key = "decoder.block." + layers + ".layer.1.EncDecAttention.o.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif op_name == "mlp": if libai_cfg.get("model_type") == "mt5": if keys[op_idx + 1] == "wi_0": new_key = "decoder.block." + layers + ".layer.2.DenseReluDense.wi_0.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) if keys[op_idx + 1] == "wi_1": new_key = "decoder.block." + layers + ".layer.2.DenseReluDense.wi_1.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) if keys[op_idx + 1] == "wo": new_key = "decoder.block." + layers + ".layer.2.DenseReluDense.wo.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) elif libai_cfg.get("model_type") == "t5": if keys[op_idx + 1] == "dense_h_to_4h": new_key = "decoder.block." + layers + ".layer.2.DenseReluDense.wi.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) if keys[op_idx + 1] == "dense_4h_to_h": new_key = "decoder.block." + layers + ".layer.2.DenseReluDense.wo.weight" torch_state_dict[new_key] = convert_tensor(oneflow_state_dict.pop(key)) return torch_state_dict if __name__ == "__main__": args = parse_args() oneflow_state_dict_path = args.oneflow_state_dict_path config_path = args.config_path save_path = args.save_path training_config = LazyConfig.load(config_path) model_config = training_config.model.cfg torch_state_dict = convert_state_dict(oneflow_state_dict_path, model_config) torch.save(torch_state_dict, save_path)
10,611
47.678899
99
py
libai
libai-main/projects/T5/utils/mask.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow from libai.utils import distributed as dist class ExtendedMask(flow.nn.Module): def forward(self, x, input_tensor=None, is_decoder=False): if x.dim() == 3: extended_mask = x[:, None, :, :] elif x.dim() == 2: if is_decoder: extended_mask = self.create_extended_mask_for_decoder(x, input_tensor) else: extended_mask = x[:, None, None, :] return extended_mask def create_extended_mask_for_decoder(self, x, input_tensor): batch_size, seq_len = input_tensor.size() seq_ids = flow.arange( seq_len, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=x.placement, ) causal_mask = ( seq_ids[None, None, :].repeat(batch_size, seq_len, 1) <= seq_ids[None, :, None] ) causal_mask = causal_mask.to(x.dtype) causal_mask = causal_mask.to_global(sbp=x.sbp) if causal_mask.shape[1] < x.shape[1]: prefix_seq_len = x.shape[1] - causal_mask.shape[1] ones = flow.ones( (batch_size, seq_len, prefix_seq_len), dtype=causal_mask.dtype, sbp=causal_mask.sbp, placement=causal_mask.placement, ) causal_mask = flow.cat( [ ones, causal_mask, ], dim=-1, ) extended_mask = causal_mask[:, None, :, :] * x[:, None, None, :] return extended_mask
2,225
33.78125
91
py
libai
libai-main/projects/MagicPrompt/gpt2.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow from oneflow import nn from oneflow.nn import init from libai.config import configurable from libai.inference.generator.generation_utils import Generator from libai.layers import Embedding, LayerNorm, LMLogits, VocabEmbedding from libai.layers.attention import AttnMaskType from libai.models.gpt_model import GPTLoss from libai.models.utils import GPT2LoaderHuggerFace, init_method_normal, scaled_init_method_normal from libai.utils import distributed as dist from projects.MagicPrompt.layers.transformer_layer import TransformerLayer class GPTModel(nn.Module, Generator): """GPT-2 language model. The output of the forward method is logits. Args: hidden_layers (int): The number of ``TransformerLayer`` in the gpt model. vocab_size (int): The size of vocabulary file. hidden_size (int): The size of hidden states. ffn_hidden_size (int): The size of intermediate layer in feed-forward network for each ``TransformerLayer``. num_attention_heads (int): The number of attention heads for each attention layer of ``TransformerLayer``. max_seq_length (int, optional): Max sequence length of input, defines the shape of Position Embeddings in GPTEmebedding. Defaults to 1024. embedding_dropout_prob (float, optional): The dropout ratio for the output of GPTEmbedding Layer. Defaults to 0.0. attention_dropout_prob (float, optional): The dropout ratio for the output of each attention layer in ``TransformerLayer``. Defaults to 0.0. output_dropout_prob (float, optional): The dropout ratio for the output for each TransformerLayer. Defaults to 0.0. layernorm_epsilon (float, optional): The epsilon of LayerNorm layer. Defaults to 1e-5. initializer_range (float, optional): Sigma of the normal distribution in the initialization method. Defaults to 0.02. use_scaled_init_for_output_weights (bool, optional): Defaults to ``True``. bias_gelu_fusion (bool, optional): Whether or not to fuse the computing of bias and gelu. Defaults to ``False``. bias_dropout_fusion (bool, optional): Whether or not to fuse the computing of dropout and bias. Defaults to ``False``. scale_mask_softmax_fusion (bool, optional): Whether to fuse the computing of mask and softmax in attention layers. Defaults to ``False``. apply_query_key_layer_scaling (bool, optional): Whether or not to use layer index related scaling in computing attention scores. If ``True``, the scaling factor equals to sqrt(d) * (layer_index + 1). Defaults to ``False``. apply_residual_post_layernorm (bool, optional): If set ``True``, use original BERT residual connection ordering otherwise use Megatron BERT residual connection which is more stable when scaling model size introduced in https://arxiv.org/pdf/1909.08053.pdf. Default: ``False``. amp_enabled (bool, optional): Whether or not to set fp16 for embedding weight in T5 model. Defaults to ``False``. """ @configurable def __init__( self, hidden_layers, vocab_size, hidden_size, ffn_hidden_size, num_attention_heads, max_seq_length=1024, embedding_dropout_prob=0.0, attention_dropout_prob=0.0, output_dropout_prob=0.0, layernorm_epsilon=1e-5, initializer_range=0.02, use_scaled_init_for_output_weights=True, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=False, apply_query_key_layer_scaling=False, apply_residual_post_layernorm=False, amp_enabled=False, cfg=None, ): super().__init__() self.cfg = cfg init_method = init_method_normal(sigma=initializer_range) if use_scaled_init_for_output_weights: output_layer_init_method = scaled_init_method_normal(initializer_range, hidden_layers) else: output_layer_init_method = init_method self.embeddings = GPTEmbedding( vocab_size, hidden_size, max_seq_length, init_method=init_method, embedding_dropout_prob=embedding_dropout_prob, amp_enabled=amp_enabled, ) self.transformer = Transformer( hidden_layers, hidden_size, ffn_hidden_size, num_attention_heads, attention_dropout_prob=attention_dropout_prob, output_dropout_prob=output_dropout_prob, layernorm_epsilon=layernorm_epsilon, init_method=init_method, output_layer_init_method=output_layer_init_method, bias_gelu_fusion=bias_gelu_fusion, bias_dropout_fusion=bias_dropout_fusion, scale_mask_softmax_fusion=scale_mask_softmax_fusion, apply_query_key_layer_scaling=apply_query_key_layer_scaling, apply_residual_post_layernorm=apply_residual_post_layernorm, set_cache=self.set_cache, ) self.past_key_values = [None] * hidden_layers self.past_length = 0 self.lm_head = LMLogits(vocab_size, bias=False) @classmethod def from_config(cls, cfg): return { "hidden_layers": cfg.hidden_layers, "vocab_size": cfg.vocab_size, "hidden_size": cfg.hidden_size, "ffn_hidden_size": cfg.ffn_hidden_size, "num_attention_heads": cfg.num_attention_heads, "max_seq_length": cfg.max_seq_length, "embedding_dropout_prob": cfg.embedding_dropout_prob, "attention_dropout_prob": cfg.attention_dropout_prob, "output_dropout_prob": cfg.output_dropout_prob, "layernorm_epsilon": cfg.layernorm_epsilon, "initializer_range": cfg.initializer_range, "use_scaled_init_for_output_weights": cfg.use_scaled_init_for_output_weights, "bias_gelu_fusion": cfg.bias_gelu_fusion, "bias_dropout_fusion": cfg.bias_dropout_fusion, "scale_mask_softmax_fusion": cfg.scale_mask_softmax_fusion, "apply_query_key_layer_scaling": cfg.apply_query_key_layer_scaling, "apply_residual_post_layernorm": cfg.apply_residual_post_layernorm, "amp_enabled": cfg.amp_enabled, "cfg": cfg, } def forward(self, input_ids, use_cache=False): """ Args: input_ids (flow.LongTensor): Indices of input sequence tokens in vocabulary. Returns: flow.Tensor: logits """ input_ids = input_ids.to_global(placement=dist.get_layer_placement(0)) if use_cache and self.past_key_values[0] is not None: self.past_length = self.past_key_values[0][0].size(-2) else: self.past_length = 0 input_embeds = self.embeddings(input_ids, self.past_length) transformer_output = self.transformer( input_embeds, attention_mask=None, past_key_values=self.past_key_values, use_cache=use_cache, ) logits = self.lm_head(transformer_output, self.embeddings.token_embeddings.weight) return {"logits": logits} def set_cache(self, past_key_values): self.past_length = 0 if past_key_values is None else past_key_values[0][0].shape[2] if past_key_values is None: past_key_values = [None] * self.cfg.hidden_layers assert len(past_key_values) == self.cfg.hidden_layers, ( f"past_key_values's length {len(past_key_values)} doesn't match " f"num_layers:' {self.cfg.hidden_layers}" ) self.past_key_values = past_key_values def _reorder_cache(self, beam_idx): past_key_values = self.past_key_values return tuple( tuple( past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past ) for layer_past in past_key_values ) def prepare_inputs_for_generation( self, input_ids, past=None, use_cache=None, ): if past is not None: input_ids = input_ids[:, -1:] self.past_key_values = past return {"input_ids": input_ids, "use_cache": use_cache} class GPTEmbedding(nn.Module): def __init__( self, vocab_size, hidden_size, max_seq_length, init_method=init.xavier_normal_, embedding_dropout_prob=0.0, amp_enabled=False, ): super().__init__() self.token_embeddings = VocabEmbedding( vocab_size, hidden_size, init_method=init_method, amp_enabled=amp_enabled ) self.position_embeddings = Embedding( max_seq_length, hidden_size, init_method=init_method, amp_enabled=amp_enabled ) self.dropout = nn.Dropout(embedding_dropout_prob) self.position_ids = flow.arange( max_seq_length, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ).unsqueeze(0) def forward(self, input_ids, past_length=0): bsz, seq_length = input_ids.size() position_ids = self.position_ids[:, past_length : past_length + seq_length] position_ids = position_ids.expand_as(input_ids).to_global(sbp=input_ids.sbp) token_embeds = self.token_embeddings(input_ids) position_embeds = self.position_embeddings(position_ids) input_embeds = token_embeds + position_embeds input_embeds = self.dropout(input_embeds) return input_embeds class Transformer(nn.Module): def __init__( self, hidden_layers, hidden_size, ffn_hidden_size, num_attention_heads, attention_dropout_prob=0.0, output_dropout_prob=0.0, layernorm_epsilon=1e-5, init_method=init.xavier_normal_, output_layer_init_method=None, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=False, apply_query_key_layer_scaling=False, apply_residual_post_layernorm=False, set_cache=None, ): super().__init__() self.hidden_layers = hidden_layers self.set_cache = set_cache def build_layer(layer_number): return TransformerLayer( hidden_size, ffn_hidden_size, num_attention_heads, attention_dropout_prob=attention_dropout_prob, output_dropout_prob=output_dropout_prob, layernorm_epsilon=layernorm_epsilon, init_method=init_method, output_layer_init_method=output_layer_init_method, bias_gelu_fusion=bias_gelu_fusion, bias_dropout_fusion=bias_dropout_fusion, scale_mask_softmax_fusion=scale_mask_softmax_fusion, apply_query_key_layer_scaling=apply_query_key_layer_scaling, apply_residual_post_layernorm=apply_residual_post_layernorm, attn_mask_type=AttnMaskType.causal, layer_idx=layer_number, ) self.layers = nn.ModuleList([build_layer(i) for i in range(self.hidden_layers)]) self.layernorm_f = LayerNorm(hidden_size, eps=layernorm_epsilon, layer_idx=-1) def forward(self, hidden_states, attention_mask, past_key_values=None, use_cache=False): if use_cache: presents = [] for layer, past_key_value in zip(self.layers, past_key_values): hidden_states = layer( hidden_states, attention_mask, past_key_value=past_key_value, use_cache=use_cache, ) if use_cache: hidden_states, present = hidden_states presents.append(present) output = self.layernorm_f(hidden_states) if use_cache: self.set_cache(presents) return output class GPTForPreTraining(nn.Module): """ GPT Model with classification head on top. """ def __init__(self, cfg) -> None: super().__init__() if cfg.pretrained_model_path is not None: loader = GPT2LoaderHuggerFace(GPTModel, cfg, cfg.pretrained_model_path) self.GPT_model = loader.load() else: self.GPT_model = GPTModel(cfg) self.loss_func = GPTLoss() def forward( self, input_ids, labels=None, ): """ Args: input_ids (flow.LongTensor): Indices of input sequence tokens in vocabulary. labels (flow.LongTensor, optional): Labels for computing language modeling loss. None for evaluating. Defaults to None. Returns: dict: A dict containing :code:`loss_value` or :code:`logits` depending on training or evaluation. :code:`{"masked_lm_loss": loss_value}` when training, :code:`{"prediction_scores": logits}` when evaluating. """ logits = self.GPT_model(input_ids)["logits"] if labels is not None: lm_loss = self.loss_func(logits, labels) return lm_loss else: return {"prediction_scores": logits} @staticmethod def set_pipeline_stage_id(model: nn.Module): dist_utils = dist.get_dist_util() if hasattr(model.GPT_model.transformer.layernorm_f, "config"): # Old API in OneFlow 0.8 for module_block in model.modules(): if isinstance(module_block.origin, (GPTEmbedding)): module_block.config.set_stage( dist_utils.get_layer_stage_id(0), dist.get_layer_placement(0) ) elif isinstance(module_block.origin, TransformerLayer): module_block.config.set_stage( dist_utils.get_layer_stage_id(module_block.layer_idx), dist.get_layer_placement(module_block.layer_idx), ) elif isinstance(module_block.origin, (LMLogits, GPTLoss)): module_block.config.set_stage( dist_utils.get_layer_stage_id(-1), dist.get_layer_placement(-1) ) model.GPT_model.transformer.layernorm_f.config.set_stage( dist_utils.get_layer_stage_id(-1), dist.get_layer_placement(-1) ) else: for module_block in model.modules(): if isinstance(module_block.to(nn.Module), (GPTEmbedding)): module_block.to(nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(0), dist.get_layer_placement(0) ) elif isinstance(module_block.to(nn.Module), TransformerLayer): module_block.to(nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(module_block.layer_idx), dist.get_layer_placement(module_block.layer_idx), ) elif isinstance(module_block.to(nn.Module), (LMLogits, GPTLoss)): module_block.to(nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(-1), dist.get_layer_placement(-1) ) model.GPT_model.transformer.layernorm_f.to(nn.graph.GraphModule).set_stage( dist_utils.get_layer_stage_id(-1), dist.get_layer_placement(-1) ) @staticmethod def set_activation_checkpoint(model): for module_block in model.modules(): # Old API in OneFlow 0.8 if hasattr(module_block, "origin"): if isinstance(module_block.origin, TransformerLayer): module_block.config.activation_checkpointing = True else: if isinstance(module_block.to(nn.Module), TransformerLayer): module_block.to(nn.graph.GraphModule).activation_checkpointing = True
17,127
38.740139
100
py
libai
libai-main/projects/MagicPrompt/pipeline.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. from libai.inference.basic import BasePipeline from libai.utils import distributed as dist class TextGenerationPipeline(BasePipeline): def load_pretrain_weight(self, libai_cfg_model, model_path, mode="huggingface"): """load pretrained model. Args: libai_cfg_model (libai.models): Lazy config Model in Libai, you can import it by `from libai.config.configs.common.models.bert import pretrain_model as libai_cfg_model` model_path (str): The directory path of pretrained model, """ if mode == "huggingface": from libai.models.utils import GPT2LoaderHuggerFace model_loader = GPT2LoaderHuggerFace( libai_cfg_model, libai_cfg_model.cfg, model_path, ) model = model_loader.load() model.eval() return model elif mode == "libai": from libai.models.utils import GPT2LoaderLiBai model_loader = GPT2LoaderLiBai( libai_cfg_model, libai_cfg_model.cfg, model_path, ) model = model_loader.load() model.eval() return model elif mode == "random": from libai.engine import DefaultTrainer return DefaultTrainer.build_model(self.cfg) else: raise NotImplementedError def _parse_parameters(self, **pipeline_parameters): preprocess_params = {} forward_params = {**pipeline_parameters} postprocess_params = {} return preprocess_params, forward_params, postprocess_params def preprocess(self, inputs, **kwargs) -> dict: # tokenizer encoder if self.tokenizer.pad_token is None: self.tokenizer.pad_token = "[PAD]" inputs = self.tokenizer(inputs, return_tensors="of", padding=True) inputs = { "input_ids": inputs.input_ids, } return inputs def forward(self, inputs, **kwargs) -> dict: outputs = self.model.generate(inputs["input_ids"], max_length=50, **kwargs) return {"return_ids": outputs} def postprocess(self, model_output_dict, **kwargs) -> dict: return_ids = model_output_dict["return_ids"] records = [ {"generated_text": self.tokenizer.decode(return_ids[i], skip_special_tokens=True)} for i in range(return_ids.size(0)) ] return records if __name__ == "__main__": pipeline = TextGenerationPipeline( "projects/MagicPrompt/configs/gpt2_inference.py", data_parallel=1, tensor_parallel=1, pipeline_parallel=1, # pipeline_stage_id=[0] * 6 + [1] * 6, # pipeline_num_layers=12, model_path="/path/to/oneflow-model", mode="libai", ) text = ["a dog", "a cute pig", "a cute girl"] output = pipeline(inputs=text) if dist.is_main_process(): print(output)
3,639
32.703704
94
py
libai
libai-main/projects/MagicPrompt/datasets/datasets.py
import os def convert_txt2json(file_path): """Store the dataset in loose JSON format file, you can refer: https://libai.readthedocs.io/en/latest/tutorials/basics/Preprocessing_Dataset.html """ filename, ext = os.path.splitext(file_path) filename = filename.split("/")[-1] with open(file_path) as f: lines = f.readlines() print(len(lines)) target_file = "/home/xiezipeng/libai/projects/MagicPrompt/" + filename + "_magicprompy.txt" with open(target_file, "w", encoding="utf-8") as f: for line in lines: line = "{" + '"' + "text" + '"' + ": " + '"' + line.strip() + '"' + "}" + "\n" f.write(line) os.rename(target_file, target_file[:-4] + ".json") if __name__ == "__main__": convert_txt2json("/path/to/test.txt") convert_txt2json("/path/to/train.txt")
848
31.653846
95
py
libai
libai-main/projects/MagicPrompt/layers/transformer_layer.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow.nn as nn from libai.layers import build_activation from libai.layers.droppath import DropPath from libai.layers.layer_norm import LayerNorm from libai.layers.mlp import MLP from libai.utils import distributed as dist from projects.MagicPrompt.layers.attention_layer import AttnMaskType, MultiheadAttention class TransformerLayer(nn.Module): """A single transformer layer. Transformer layer takes input with size [bsz, seq_length, hidden size] and returns an output of the same size. The input and output has same sbp sign, (S(0), B). Arguments: hidden_size: size of hidden state. ffn_hidden_size: size of feed forword neural network. num_attention_heads: number of attention heads. is_decoder: used to specify whether this is transformer encoder layer or transformer decoder layer. Default: ``False``. attention_dropout_prob: dropout probability of attention weights. output_dropout_prob: dropout probability of output. layernorm_epsilon: epsilon used in layernorm layer. Default: `1e-5`. init_method: method to initialize the input layer weights. output_layer_init_method: method to initialize the output layer weights. If None, use `init_method`. bias_gelu_fusion: whether fuse add bias and gelu. Default: ``False``. bias_dropout_fusion: whether fuse add bias and dropout. Default: ``False``. scale_mask_softmax_fusion: whether to fuse scale, mask and softmax. Default: ``False``. apply_query_key_layer_scaling: if `true`, scaling the attention score by layer index. Default: ``False``. apply_residual_post_layernorm: if ``true``, use original BERT residual connection ordering. Otherwise, use Megatron BERT residual connection which is more stable when scaling model size introduced in https://arxiv.org/pdf/1909.08053.pdf. Default: ``False``. layer_idx: the layer index, which determines the placement. """ def __init__( self, hidden_size, ffn_hidden_size, num_attention_heads, is_decoder=False, attention_dropout_prob=0.0, output_dropout_prob=0.0, drop_path_prob=0.0, layernorm_epsilon=1e-5, init_method=nn.init.xavier_normal_, output_layer_init_method=None, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=False, apply_query_key_layer_scaling=False, apply_residual_post_layernorm=False, attn_mask_type=AttnMaskType.padding, *, layer_idx=0 ): super().__init__() self.hidden_size = hidden_size self.ffn_hidden_size = ffn_hidden_size self.num_attention_heads = num_attention_heads self.attention_dropout_prob = attention_dropout_prob self.output_dropout_prob = output_dropout_prob self.layernorm_epsilon = layernorm_epsilon self.attn_mask_type = attn_mask_type self.layer_idx = layer_idx self.is_decoder = is_decoder self.bias_gelu_fusion = bias_gelu_fusion self.bias_dropout_fusion = bias_dropout_fusion self.scale_mask_softmax_fusion = scale_mask_softmax_fusion self.apply_query_key_layer_scaling = apply_query_key_layer_scaling self.apply_residual_post_layernorm = apply_residual_post_layernorm self.init_method = init_method if output_layer_init_method is None: output_layer_init_method = init_method self.output_layer_init_method = output_layer_init_method self.drop_path = DropPath(drop_path_prob) if drop_path_prob > 0.0 else nn.Identity() self.input_layernorm = LayerNorm( self.hidden_size, eps=self.layernorm_epsilon, layer_idx=self.layer_idx ) self.self_attention = self.build_attention(is_cross_attention=False) self.post_attention_layernorm = LayerNorm( self.hidden_size, eps=self.layernorm_epsilon, layer_idx=self.layer_idx ) if self.is_decoder: self.cross_attention = self.build_attention(is_cross_attention=True) self.post_cross_attention_layernorm = LayerNorm( self.hidden_size, eps=self.layernorm_epsilon, layer_idx=self.layer_idx ) self.mlp = MLP( self.hidden_size, self.ffn_hidden_size, self.output_dropout_prob, self.init_method, output_layer_init_method=self.output_layer_init_method, bias_gelu_fusion=self.bias_gelu_fusion, bias_dropout_fusion=self.bias_dropout_fusion, layer_idx=self.layer_idx, ) # use gelu_tanh activation if not bias_gelu_fusion: self.mlp.activation_func = build_activation("gelu_tanh") def forward( self, hidden_states, attention_mask=None, encoder_states=None, encoder_attention_mask=None, past_key_value=None, use_cache=False, ): """ Args: hidden_states: shape is (batch_size, seq_length, hidden_size), sbp signature is (S(0), B). attention_mask: the combination of key padding mask and casual mask of hidden states with shape (batch_size, 1, seq_length, seq_length) and the sbp signature is (S(0), B), encoder_states: encoder output with shape (batch_size, seq_length, hidden_size) and the sbp signature is (S(0), B), which will be used in cross attention. encoder_attention_mask: key padding mask of encoder states with shape (batch_size, 1, seq_length, seq_length) and the sbp signature is (S(0), B). past_key_value: tuple of key and value, each shape is (seq_length, bsz, num_heads, head_size), For decoder layer, the past_key_value contains the states both from self attention and cross attention. use_cache: it will be set to `True` when the model is in the inference phase and used for incremental decoding. """ hidden_states = hidden_states.to_global(placement=dist.get_layer_placement(self.layer_idx)) if attention_mask is not None: attention_mask = attention_mask.to_global( placement=dist.get_layer_placement(self.layer_idx) ) if past_key_value is not None: if self.is_decoder: assert len(past_key_value) == 4 self_attn_past_key_value = past_key_value[:2] cross_attn_past_key_value = past_key_value[2:] else: self_attn_past_key_value = past_key_value cross_attn_past_key_value = None else: self_attn_past_key_value, cross_attn_past_key_value = None, None layernorm_output = self.input_layernorm(hidden_states) attention_output = self.self_attention( layernorm_output, attention_mask=attention_mask, past_key_value=self_attn_past_key_value, use_cache=use_cache, ) attention_output = self.drop_path(attention_output) if use_cache: attention_output, presents = attention_output if self.apply_residual_post_layernorm: residual = layernorm_output else: residual = hidden_states hidden_states = residual + attention_output layernorm_output = self.post_attention_layernorm(hidden_states) if self.is_decoder: attention_output = self.cross_attention( layernorm_output, encoder_states, attention_mask=encoder_attention_mask, past_key_value=cross_attn_past_key_value, use_cache=use_cache, ) if use_cache: attention_output, decoder_presents = attention_output presents += decoder_presents attention_output = self.drop_path(attention_output) if self.apply_residual_post_layernorm: residual = layernorm_output else: residual = hidden_states hidden_states = residual + attention_output layernorm_output = self.post_cross_attention_layernorm(hidden_states) mlp_output = self.mlp(layernorm_output) mlp_output = self.drop_path(mlp_output) if self.apply_residual_post_layernorm: residual = layernorm_output else: residual = hidden_states output = residual + mlp_output if use_cache: output = (output, presents) return output def build_attention(self, is_cross_attention=False): return MultiheadAttention( self.hidden_size, self.num_attention_heads, is_cross_attention=is_cross_attention, attention_dropout_prob=self.attention_dropout_prob, output_dropout_prob=self.output_dropout_prob, init_method=self.init_method, output_layer_init_method=self.output_layer_init_method, bias_dropout_fusion=self.bias_dropout_fusion, scale_mask_softmax_fusion=self.scale_mask_softmax_fusion, apply_query_key_layer_scaling=self.apply_query_key_layer_scaling, attn_mask_type=self.attn_mask_type, layer_idx=self.layer_idx, )
10,211
39.685259
99
py
libai
libai-main/projects/MagicPrompt/layers/__init__.py
0
0
0
py
libai
libai-main/projects/MagicPrompt/layers/attention_layer.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import math from typing import Tuple import oneflow as flow from oneflow import nn from libai.layers.attention import AttnMaskType from libai.layers.linear import Linear from libai.utils import distributed as dist class MultiheadAttention(nn.Module): """Multi-head attention layer, support self attention and cross attention. Args: hidden_size: size of hidden state. num_attention_heads: number of attention heads. is_cross_attention: used to specify whether it is self attention or cross attention. Defaults to False. attention_dropout_prob: dropout probability of attention weights. Defaults to 0.0. output_dropout_prob: dropout probability of output. Defaults to 0.0. init_method: method to initialize the input layer weights. Defaults to ``init.xavier_normal_``. output_layer_init_method: method to initialize the output layer weights. If None, use ``init_method``. bias_dropout_fusion: whether to fuse add bias and dropout. Defaults to False. scale_mask_softmax_fusion: whether to fuse scale, mask and softmax. Defaults to False. apply_query_key_layer_scaling: if `True`, scaling the attention score by layer index. Defaults to False. layer_idx: a layer_idx sign which determines the placements. It will be used in pipeline parallelism. Defaults to 0. """ def __init__( self, hidden_size, num_attention_heads, is_cross_attention=False, attention_dropout_prob=0.0, output_dropout_prob=0.0, init_method=nn.init.xavier_normal_, output_layer_init_method=None, bias_dropout_fusion=False, scale_mask_softmax_fusion=False, apply_query_key_layer_scaling=False, attn_mask_type=AttnMaskType.padding, *, layer_idx=0 ): super().__init__() self.hidden_size = hidden_size if output_layer_init_method is None: output_layer_init_method = init_method assert ( hidden_size % num_attention_heads == 0 ), "hidden_size must be divisible by num_attention_heads." self.num_heads = num_attention_heads self.head_size = hidden_size // num_attention_heads self.attn_mask_type = attn_mask_type self.attention_dropout_prob = attention_dropout_prob self.dropout = nn.Dropout(p=attention_dropout_prob) self.norm_factor = 1.0 / math.sqrt(float(self.head_size)) self.coeff = None if apply_query_key_layer_scaling: self.coeff = layer_idx + 1 self.norm_factor /= self.coeff self.is_cross_attention = is_cross_attention self.scale_mask_softmax_fusion = scale_mask_softmax_fusion self.bias_dropout_fusion = bias_dropout_fusion if self.bias_dropout_fusion: self.output_dropout_prob = output_dropout_prob else: self.output_dropout = nn.Dropout(p=output_dropout_prob) if self.is_cross_attention: self.query = Linear( self.hidden_size, self.hidden_size, parallel="col", init_method=init_method, layer_idx=layer_idx, ) self.key_value = Linear( self.hidden_size, self.hidden_size * 2, parallel="col", init_method=init_method, layer_idx=layer_idx, ) else: self.query_key_value = Linear( self.hidden_size, self.hidden_size * 3, parallel="col", init_method=init_method, layer_idx=layer_idx, ) self.dense = Linear( self.hidden_size, self.hidden_size, parallel="row", init_method=output_layer_init_method, skip_bias_add=self.bias_dropout_fusion, layer_idx=layer_idx, ) self.bias = flow.tril(flow.ones((1024, 1024), dtype=flow.uint8)).view(1, 1, 1024, 1024) self.bias = self.bias.to_global( sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(layer_idx), ) def forward( self, hidden_states: flow.Tensor, encoder_states: flow.Tensor = None, attention_mask: flow.Tensor = None, past_key_value: Tuple[flow.Tensor, flow.Tensor] = None, use_cache: bool = False, ): """ Args: hidden_states (flow.Tensor): shape is [bsz, tgt_len, hidden_size]. encoder_states (flow.Tensor, optional): shape is [bsz, src_len, hidden_size]. Defaults to None. attention_mask (flow.Tensor, optional): shape is [bsz, 1, tgt_len, src_len]. It should be the combination of padding mask and casual mask. It is the padding mask of source input when used with self-attention in encoder. And it is the combination of padding mask of target input and casual mask when used with self-attention in decoder. It is the padding mask of source input when used with cross-attention in decoder. Defaults to None. past_key_value (Tuple[flow.Tensor, flow.Tensor], optional): tuple of key and value, each shape is [bsz, num_heads, src_len, head_size]. Defaults to None. use_cache (bool, optional): it will be set to True, when the model is in the inference phase and used for incremental decoding. Defaults to False. """ if encoder_states is not None: encoder_states = encoder_states.to_global(placement=hidden_states.placement) if attention_mask is not None: attention_mask = attention_mask.to_global(placement=hidden_states.placement) bsz, tgt_len = hidden_states.size()[:2] if self.is_cross_attention: query = self.query(hidden_states) query = query.view(bsz, -1, self.num_heads, self.head_size) query = query.permute(0, 2, 1, 3) if past_key_value is not None: key, value = past_key_value elif encoder_states is not None: key_value = self.key_value(encoder_states) key_value = key_value.view(bsz, -1, self.num_heads, 2 * self.head_size) key_value = key_value.permute(0, 2, 1, 3) key, value = flow.chunk(key_value, chunks=2, dim=-1) else: raise ValueError( "past_key_value and encoder_states cannot be None at the same time." ) else: query_key_value = self.query_key_value(hidden_states) query_key_value = query_key_value.view(bsz, -1, self.num_heads, 3 * self.head_size) query_key_value = query_key_value.permute( 0, 2, 1, 3 ) # [bsz, num_heads, src_len, 3 * head_size] query, key, value = flow.chunk(query_key_value, chunks=3, dim=-1) if past_key_value is not None: past_key, past_value = past_key_value key = flow.cat((past_key.type_as(key), key), dim=2) value = flow.cat((past_value.type_as(value), value), dim=2) if use_cache: past_key_value = (key, value) attention_scores = flow.matmul(query, key, transpose_b=True, alpha=self.norm_factor) if not self.is_cross_attention: query_length, key_length = query.size(-2), key.size(-2) causal_mask = self.bias[:, :, key_length - query_length : key_length, :key_length].to( flow.bool ) causal_mask = causal_mask.repeat(attention_scores.size(0), 1, 1, 1) causal_mask = causal_mask.to_global(placement=attention_scores.placement) fill_value = flow.finfo(attention_scores.dtype).min if causal_mask.shape[0] == 1: mask_value = flow.ones( causal_mask.size(), dtype=attention_scores.dtype, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=attention_scores.placement, ).fill_(fill_value) else: mask_value = flow.ones( causal_mask.size(), dtype=attention_scores.dtype, sbp=dist.get_nd_sbp([flow.sbp.split(0), flow.sbp.broadcast]), placement=attention_scores.placement, ).fill_(fill_value) attention_scores = flow.where(causal_mask, attention_scores, mask_value) if attention_mask is not None: if self.scale_mask_softmax_fusion: if self.attn_mask_type == AttnMaskType.padding: attention_mask = ( attention_mask.expand_as(attention_scores) if use_cache else attention_mask ) attention_weights = flow._C.fused_scale_mask_softmax_dropout( attention_scores, attention_mask, fill_value=-10000.0, scale=self.coeff, p=self.attention_dropout_prob, )[0] else: if self.coeff is not None: attention_scores *= self.coeff attention_scores = flow.mul(attention_scores, attention_mask) attention_scores = attention_scores - 10000.0 * (1 - attention_mask) attention_weights = flow.softmax(attention_scores, dim=-1) attention_weights = self.dropout(attention_weights) else: if self.scale_mask_softmax_fusion and self.attn_mask_type == AttnMaskType.causal: attention_weights = flow._C.fused_scale_tril_softmax_mask_scale( attention_scores, p=self.attention_dropout_prob, diagonal=0, tril_scale_value=self.coeff, tril_fill_value=-10000.0, )[0] else: attention_weights = flow.softmax(attention_scores, dim=-1) attention_weights = self.dropout(attention_weights) context = flow.matmul(attention_weights, value) context = context.transpose(1, 2) output = self.dense(context.flatten(2)) if self.bias_dropout_fusion: output, bias = output output = flow._C.fused_bias_add_dropout( output, bias, p=self.output_dropout_prob, axis=output.ndim - 1 ) else: output = self.output_dropout(output) if use_cache: output = (output, past_key_value) return output def extra_repr(self) -> str: return "hidden_size={}, num_heads={}, is_cross_attention={}".format( self.hidden_size, self.num_heads, self.is_cross_attention, )
11,880
40.982332
99
py
libai
libai-main/projects/MagicPrompt/configs/gpt2_dataset.py
from libai.config import LazyCall from libai.data import build_nlp_train_val_test_loader from configs.common.data.gpt_dataset import tokenization, dataloader from libai.tokenizer import GPT2Tokenizer from libai.data.datasets import GPT2Dataset from libai.data.data_utils import get_indexed_dataset data_prefix = "/data/home/magicprompt/train/en_train_mmap_text_sentence" tokenization.tokenizer = LazyCall(GPT2Tokenizer)( vocab_file="/data/home/magicprompt/vocab.json", merges_file="/data/home/magicprompt/merges.txt", do_lower_case=True, do_chinese_wwm=True, ) tokenization.append_eod = False dataloader.train = LazyCall(build_nlp_train_val_test_loader)( dataset=[ LazyCall(GPT2Dataset)( name="gpt-2", data_prefix=data_prefix, indexed_dataset=LazyCall(get_indexed_dataset)( data_prefix=data_prefix, data_impl="mmap", skip_warmup=False, ), max_seq_length=1024, seed=1234, ), ], train_val_test_num_samples=None, # a hint for deferred assignment splits=[[949.0, 50.0, 1.0]], weights=[1.0], num_workers=4, )
1,185
30.210526
72
py
libai
libai-main/projects/MagicPrompt/configs/gpt2_training.py
from libai.config import LazyCall from libai.evaluation import PPLEvaluator from projects.MagicPrompt.configs.gpt2_inference import pretrain_model as model from projects.MagicPrompt.configs.gpt2_dataset import dataloader, tokenization from configs.common.optim import optim from libai.scheduler import WarmupExponentialLR from configs.common.train import train from configs.common.models.graph import graph vocab_file = "/data/home/magicprompt/vocab.json" merge_files = "/data/home/magicprompt/merges.txt" train_data_prefix = "/data/home/magicprompt/train/en_train_mmap_text_sentence" tokenization.tokenizer.vocab_file = vocab_file tokenization.tokenizer.merges_file = merge_files dataloader.train.dataset[0].data_prefix = train_data_prefix dataloader.train.dataset[0].indexed_dataset.data_prefix = train_data_prefix # gpt2 model config model.cfg.pretrained_model_path = None model.cfg.embedding_dropout_prob = 0.1 model.cfg.attention_dropout_prob = 0.1 model.cfg.output_dropout_prob = 0.1 model.cfg.num_attention_heads = 12 model.cfg.hidden_size = 768 model.cfg.ffn_hidden_size = 4 * 768 model.cfg.hidden_layers = 12 model.cfg.max_seq_length = 1024 model.cfg.initializer_range = 0.02 model.cfg.vocab_size = 50257 model.cfg.layernorm_epsilon = 1e-5 model.cfg.use_scaled_init_for_output_weights = True model.cfg.bias_gelu_fusion = True model.cfg.bias_dropout_fusion = True model.cfg.scale_mask_softmax_fusion = False model.cfg.apply_query_key_layer_scaling = True model.cfg.apply_residual_post_layernorm = False model.cfg.amp_enabled = True train.input_placement_device = "cpu" train.dist.pipeline_num_layers = model.cfg.hidden_layers for ds in dataloader.train.dataset: ds.max_seq_length = model.cfg.max_seq_length optim.lr = 5.0e-05 train.update( dict( output_dir="projects/MagicPrompt/oneflow_magicprompt", train_micro_batch_size=4, test_micro_batch_size=4, train_epoch=33, train_iter=10000, log_period=50, amp=dict(enabled=True), warmup_ratio=0, checkpointer=dict(period=8000, max_to_keep=20), dist=dict( data_parallel_size=1, tensor_parallel_size=1, pipeline_parallel_size=1, # pipeline_num_layers=model.cfg.hidden_layers, ), scheduler=LazyCall(WarmupExponentialLR)( warmup_factor=0.0, gamma=1.0, warmup_method="linear", warmup_iter=0.0, ), evaluation=dict( enabled=True, evaluator=LazyCall(PPLEvaluator)(), eval_iter=250, eval_period=4000, ), rdma_enabled=False, ) )
2,667
30.761905
79
py
libai
libai-main/projects/MagicPrompt/configs/gpt2_inference.py
from configs.common.models.gpt import cfg from libai.config import LazyCall from projects.mock_transformers import mock_tokenization from projects.MagicPrompt.gpt2 import GPTModel, GPTForPreTraining from configs.common.data.gpt_dataset import tokenization from configs.common.train import train cfg.update( # Model embedding_dropout_prob=0, attention_dropout_prob=0, output_dropout_prob=0, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=False, apply_query_key_layer_scaling=False, apply_residual_post_layernorm=False, amp_enabled=False, num_attention_heads=12, hidden_size=768, ffn_hidden_size=4 * 768, hidden_layers=12, max_seq_length=1024, initializer_range=0.02, vocab_size=50304, # Inference is_encoder_decoder=False, max_length=20, min_length=0, do_sample=False, early_stopping=False, num_beams=1, num_beam_groups=1, diversity_penalty=0.0, temperature=1.0, top_k=50, top_p=1.0, typical_p=1.0, repetition_penalty=1.0, length_penalty=1.0, no_repeat_ngram_size=0, encoder_no_repeat_ngram_size=0, num_return_sequences=1, chunk_size_feed_forward=0, output_scores=False, forced_bos_token_id=None, forced_eos_token_id=None, remove_invalid_values=False, exponential_decay_length_penalty=None, use_cache=True, # Tokenizer pad_token_id=0, eos_token_id=50256, bos_token_id=50256, sep_token_id=None, decoder_start_token_id=None, # train pretrained_model_path="/data/home/magicprompt", ) model = LazyCall(GPTModel)(cfg=cfg) pretrain_model = LazyCall(GPTForPreTraining)(cfg=cfg) tokenization.tokenizer = LazyCall(mock_tokenization.GPT2Tokenizer)( vocab_file="/data/home/magicprompt/vocab.json", merges_file="/data/home/magicprompt/merges.txt", )
1,879
26.246377
67
py
libai
libai-main/projects/Stable_Diffusion/modeling.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow from diffusers import AutoencoderKL, DDPMScheduler, UNet2DConditionModel from diffusers.loaders import AttnProcsLayers from diffusers.models.cross_attention import LoRACrossAttnProcessor from oneflow import nn from oneflow.nn import functional as F from transformers import CLIPTextModel, CLIPTokenizer from projects.mock_transformers import init_env # noqa LoRACrossAttnProcessor.forward = LoRACrossAttnProcessor.__call__ class StableDiffusion(nn.Module): def __init__( self, model_path, train_vae=False, train_text_encoder=False, train_with_lora=False, ): super().__init__() self.model_path = model_path self.tokenizer = CLIPTokenizer.from_pretrained(model_path, subfolder="tokenizer") self.text_encoder = CLIPTextModel.from_pretrained(model_path, subfolder="text_encoder") self.vae = AutoencoderKL.from_pretrained(model_path, subfolder="vae") self.unet = UNet2DConditionModel.from_pretrained(model_path, subfolder="unet") self.noise_scheduler = DDPMScheduler.from_pretrained(model_path, subfolder="scheduler") for name in self.noise_scheduler.__dict__.keys(): if flow.is_tensor(getattr(self.noise_scheduler, name)): setattr( self.noise_scheduler, name, getattr(self.noise_scheduler, name).to_global( sbp=flow.sbp.broadcast, placement=flow.env.all_device_placement("cuda") ), ) if not train_with_lora: if not train_vae: self.vae.requires_grad_(False) if not train_text_encoder: self.text_encoder.requires_grad_(False) else: self.vae.requires_grad_(False) self.text_encoder.requires_grad_(False) self.unet.requires_grad_(False) # Set correct lora layers lora_attn_procs = {} for name in self.unet.attn_processors.keys(): cross_attention_dim = ( None if name.endswith("attn1.processor") else self.unet.config.cross_attention_dim ) if name.startswith("mid_block"): hidden_size = self.unet.config.block_out_channels[-1] elif name.startswith("up_blocks"): block_id = int(name[len("up_blocks.")]) hidden_size = list(reversed(self.unet.config.block_out_channels))[block_id] elif name.startswith("down_blocks"): block_id = int(name[len("down_blocks.")]) hidden_size = self.unet.config.block_out_channels[block_id] lora_attn_procs[name] = LoRACrossAttnProcessor( hidden_size=hidden_size, cross_attention_dim=cross_attention_dim ) self.unet.set_attn_processor(lora_attn_procs) self.lora_layers = AttnProcsLayers(self.unet.attn_processors) def forward(self, pixel_values, input_ids): from oneflow.utils.global_view import global_mode placement_sbp_dict = dict( placement=flow.env.all_device_placement("cuda"), sbp=flow.sbp.split(0), ) with global_mode(True, **placement_sbp_dict): latents = self.vae.encode(pixel_values).latent_dist.sample() latents = latents * 0.18215 # Sample noise that we'll add to the latents noise = flow.randn( latents.shape, sbp=latents.sbp, placement=latents.placement, dtype=self.unet.dtype ).to(latents.device) bsz = latents.shape[0] # Sample a random timestep for each image timesteps = flow.randint( 0, self.noise_scheduler.config.num_train_timesteps, (bsz,), sbp=latents.sbp, placement=latents.placement, dtype=flow.long, ) # Add noise to the latents according to the noise magnitude at each timestep # (this is the forward diffusion process) noisy_latents = self.noise_scheduler.add_noise(latents, noise, timesteps) noisy_latents = noisy_latents.to(dtype=self.unet.dtype) # Get the text embedding for conditioning encoder_hidden_states = self.text_encoder(input_ids)[0] # Predict the noise residual noise_pred = self.unet(noisy_latents, timesteps, encoder_hidden_states).sample if self.noise_scheduler.config.prediction_type == "epsilon": target = noise elif self.noise_scheduler.config.prediction_type == "v_prediction": target = self.noise_scheduler.get_velocity(latents, noise, timesteps) else: raise ValueError( f"Unknown prediction type {self.noise_scheduler.config.prediction_type}" ) loss = F.mse_loss(noise_pred.float(), target.float(), reduction="mean") return {"loss": loss} @staticmethod def set_activation_checkpoint(model): from diffusers.models.unet_2d_blocks import ( DualTransformer2DModel, ResnetBlock2D, Transformer2DModel, ) from transformers.models.clip.modeling_clip import CLIPEncoder for module_block in model.modules(): prefix_name = module_block.to(nn.graph.GraphModule).name_prefix # unset vae checkpointing if prefix_name.startswith("model.vae"): continue # set clip checkpointing elif isinstance(module_block.to(nn.Module), CLIPEncoder): module_block.to(nn.graph.GraphModule).activation_checkpointing = True # set unet checkpointing elif isinstance( module_block.to(nn.Module), (ResnetBlock2D, DualTransformer2DModel, Transformer2DModel), ): module_block.to(nn.graph.GraphModule).activation_checkpointing = True
6,819
41.360248
98
py
libai
libai-main/projects/Stable_Diffusion/dataset.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os from pathlib import Path import numpy as np import oneflow as flow from flowvision import transforms from oneflow.utils.data import Dataset from PIL import Image from libai.data.structures import DistTensorData, Instance class DreamBoothDataset(Dataset): """ A dataset to prepare the instance and class images with the prompts for fine-tuning the model. It pre-processes the images and the tokenizes prompts. """ def __init__( self, instance_data_root, instance_prompt, tokenizer, tokenizer_pretrained_folder=None, class_data_root=None, class_prompt=None, size=512, center_crop=False, ): self.size = size self.center_crop = center_crop self.tokenizer = tokenizer if tokenizer_pretrained_folder: self.tokenizer = self.tokenizer.from_pretrained( tokenizer_pretrained_folder[0], subfolder=tokenizer_pretrained_folder[1] ) self.instance_data_root = Path(instance_data_root) if not self.instance_data_root.exists(): raise ValueError("Instance images root doesn't exists.") self.instance_images_path = list(Path(instance_data_root).iterdir()) self.num_instance_images = len(self.instance_images_path) self.instance_prompt = instance_prompt self._length = self.num_instance_images if class_data_root is not None: self.class_data_root = Path(class_data_root) self.class_data_root.mkdir(parents=True, exist_ok=True) self.class_images_path = list(self.class_data_root.iterdir()) self.num_class_images = len(self.class_images_path) self._length = max(self.num_class_images, self.num_instance_images) self.class_prompt = class_prompt else: self.class_data_root = None self.image_transforms = transforms.Compose( [ transforms.Resize(size, interpolation=transforms.InterpolationMode.BILINEAR), transforms.CenterCrop(size) if center_crop else transforms.RandomCrop(size), transforms.ToTensor(), transforms.Normalize([0.5], [0.5]), ] ) def __len__(self): return self._length def __getitem__(self, index): if self.class_data_root and np.random.rand() > 0.5: class_image = Image.open(self.class_images_path[index % self.num_class_images]) if not class_image.mode == "RGB": class_image = class_image.convert("RGB") instance_images = self.image_transforms(class_image) input_ids = self.tokenizer( self.class_prompt, truncation=True, padding="max_length", max_length=self.tokenizer.model_max_length, return_tensors="np", ).input_ids else: instance_image = Image.open(self.instance_images_path[index % self.num_instance_images]) if not instance_image.mode == "RGB": instance_image = instance_image.convert("RGB") instance_images = self.image_transforms(instance_image) input_ids = self.tokenizer( self.instance_prompt, truncation=True, padding="max_length", max_length=self.tokenizer.model_max_length, return_tensors="np", ).input_ids return Instance( pixel_values=DistTensorData(instance_images.to(dtype=flow.float32)), input_ids=DistTensorData(flow.tensor(input_ids[0])), ) class PromptDataset(Dataset): "A simple dataset to prepare the prompts to generate class images on multiple GPUs." def __init__(self, prompt, num_samples): self.prompt = prompt self.num_samples = num_samples def __len__(self): return self.num_samples def __getitem__(self, index): example = {} example["prompt"] = self.prompt example["index"] = index return example class TXTDataset(Dataset): def __init__( self, foloder_name, tokenizer, tokenizer_pretrained_folder=None, thres=0.2, size=512, center_crop=False, ): print(f"Loading folder data from {foloder_name}.") self.image_paths = [] self.tokenizer = tokenizer if tokenizer_pretrained_folder: self.tokenizer = self.tokenizer.from_pretrained( tokenizer_pretrained_folder[0], subfolder=tokenizer_pretrained_folder[1] ) for each_file in os.listdir(foloder_name): if each_file.endswith(".jpg"): self.image_paths.append(os.path.join(foloder_name, each_file)) self.image_transforms = transforms.Compose( [ transforms.Resize(size, interpolation=transforms.InterpolationMode.BILINEAR), transforms.CenterCrop(size) if center_crop else transforms.RandomCrop(size), transforms.ToTensor(), transforms.Normalize([0.5], [0.5]), ] ) print("Done loading data. Len of images:", len(self.image_paths)) def __len__(self): return len(self.image_paths) def __getitem__(self, idx): img_path = str(self.image_paths[idx]) instance_image = Image.open(img_path) if not instance_image.mode == "RGB": instance_image = instance_image.convert("RGB") instance_images = self.image_transforms(instance_image) caption_path = img_path.replace(".jpg", ".txt") with open(caption_path, "r") as f: caption = f.read() input_ids = self.tokenizer( caption, padding="max_length", truncation=True, max_length=self.tokenizer.model_max_length, return_tensors="np", ).input_ids return Instance( pixel_values=DistTensorData(instance_images.to(dtype=flow.float32)), input_ids=DistTensorData(flow.tensor(input_ids[0])), )
6,839
35.190476
100
py
libai
libai-main/projects/Stable_Diffusion/generate_prior_image.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import argparse import hashlib from pathlib import Path import oneflow as flow from diffusers import OneFlowStableDiffusionPipeline from tqdm import tqdm from projects.Stable_Diffusion.dataset import PromptDataset def parse_args(input_args=None): parser = argparse.ArgumentParser(description="Simple example of a training script.") parser.add_argument( "--pretrained_model_name_or_path", type=str, default=None, required=True, help="Path to pretrained model or model identifier from huggingface.co/models.", ) parser.add_argument( "--revision", type=str, default=None, required=False, help=( "Revision of pretrained model identifier from huggingface.co/models. " "Trainable model components should be float32 precision." ), ) parser.add_argument( "--class_data_dir", type=str, default=None, required=True, help="A folder containing the training data of class images.", ) parser.add_argument( "--class_prompt", type=str, default=None, required=True, help="The prompt to specify images in the same class as provided instance images.", ) parser.add_argument( "--num_class_images", type=int, default=100, required=False, help=( "Minimal class images for prior preservation loss. " "If there are not enough images already present in " "class_data_dir, additional images will be sampled with class_prompt." ), ) parser.add_argument( "--sample_batch_size", type=int, default=4, help="Batch size (per device) for sampling images.", ) parser.add_argument( "--prior_generation_precision", type=str, default="fp16", choices=["no", "fp32", "fp16", "bf16"], help=( "Choose prior generation precision between fp32, fp16 and bf16 (bfloat16)." " Bf16 requires PyTorch >=1.10.and an Nvidia Ampere GPU. " " Default to fp16 if a GPU is available else fp32." ), ) if input_args is not None: args = parser.parse_args(input_args) else: args = parser.parse_args() return args def main(args): class_images_dir = Path(args.class_data_dir) if not class_images_dir.exists(): class_images_dir.mkdir(parents=True) cur_class_images = len(list(class_images_dir.iterdir())) if args.prior_generation_precision == "fp32": torch_dtype = flow.float32 elif args.prior_generation_precision == "fp16": torch_dtype = flow.float16 elif args.prior_generation_precision == "bf16": torch_dtype = flow.bfloat16 if cur_class_images < args.num_class_images: pipeline = OneFlowStableDiffusionPipeline.from_pretrained( args.pretrained_model_name_or_path, use_auth_token=True, revision=args.revision, torch_dtype=torch_dtype, ).to("cuda") pipeline.set_progress_bar_config(disable=True) num_new_images = args.num_class_images - cur_class_images print(f"Number of class images to sample: {num_new_images}.") sample_dataset = PromptDataset(args.class_prompt, num_new_images) sample_dataloader = flow.utils.data.DataLoader( sample_dataset, batch_size=args.sample_batch_size ) for example in tqdm(sample_dataloader, desc="Generating class images"): images = pipeline(example["prompt"]).images for i, image in enumerate(images): hash_image = hashlib.sha1(image.tobytes()).hexdigest() image_filename = ( class_images_dir / f"{example['index'][i] + cur_class_images}-{hash_image}.jpg" ) image.save(image_filename) del pipeline if flow.cuda.is_available(): flow.cuda.empty_cache() return if __name__ == "__main__": args = parse_args() main(args)
4,744
31.951389
99
py
libai
libai-main/projects/Stable_Diffusion/train_net.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import logging import os import random import sys import numpy as np import oneflow as flow sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), os.path.pardir))) from onediff import OneFlowStableDiffusionPipeline # noqa from libai.config import LazyConfig, default_argument_parser, try_get_key # noqa from libai.engine import DefaultTrainer, default_setup, hooks # noqa from libai.engine.trainer import HookBase # noqa from libai.utils import distributed as dist # noqa from libai.utils.checkpoint import Checkpointer # noqa logger = logging.getLogger("libai." + __name__) class SdCheckpointer(HookBase): def __init__( self, model: flow.nn.Module, save_path: str, ) -> None: self._model = model self._save_path = save_path def after_train(self): def model_to_local(model): model.zero_grad(set_to_none=True) model = model.to_global( sbp=flow.sbp.broadcast, placement=flow.env.all_device_placement("cpu") ) return model.to_local() if hasattr(self._model, "lora_layers"): unet = model_to_local(self._model.unet) save_path = os.path.join(self._save_path, "model_sd_for_inference") logger.info(f"saving stable diffusion model to {save_path}") if dist.is_main_process(): unet.save_attn_procs(save_path) else: pipeline = OneFlowStableDiffusionPipeline.from_pretrained( self._model.model_path, tokenizer=self._model.tokenizer, text_encoder=model_to_local(self._model.text_encoder), vae=model_to_local(self._model.vae), unet=model_to_local(self._model.unet), ) save_path = os.path.join(self._save_path, "model_sd_for_inference") logger.info(f"saving stable diffusion model to {save_path}") if dist.is_main_process(): pipeline.save_pretrained(save_path) class Trainer(DefaultTrainer): def build_hooks(self): """ Build a list of default hooks, including timing, evaluation, checkpointing, lr scheduling, precise BN, writing events. Returns: list[HookBase]: """ ret = [ hooks.IterationTimer(), hooks.LRScheduler(), # for beauty lr scheduler printer in `nn.Graph` mode SdCheckpointer(self.model, self.cfg.train.output_dir), ] if not try_get_key(self.cfg, "model.train_with_lora", default=False): ret.append( hooks.PeriodicCheckpointer(self.checkpointer, self.cfg.train.checkpointer.period), ) if self.cfg.train.evaluation.enabled: assert self.cfg.train.evaluation.eval_iter > 0, "run_iter must be positive number" def test_and_save_results(): model = self.graph_eval if self.cfg.graph.enabled else self.model self._last_eval_results = self.test(self.cfg, self.test_loader, model) return self._last_eval_results if dist.is_main_process(): # run writers in the end, so that evaluation metrics are written ret.append(hooks.PeriodicWriter(self.build_writers(), self.cfg.train.log_period)) return ret def main(args): cfg = LazyConfig.load(args.config_file) cfg = LazyConfig.apply_overrides(cfg, args.opts) default_setup(cfg, args) seed_for_rank = cfg.train.seed + flow.env.get_rank() flow.manual_seed(seed_for_rank) flow.cuda.manual_seed(seed_for_rank) np.random.seed(seed_for_rank) random.seed(seed_for_rank) if args.fast_dev_run: cfg.train.train_epoch = 0 cfg.train.train_iter = 20 cfg.train.evaluation.eval_period = 10 cfg.train.log_period = 1 if args.eval_only: tokenizer = None if try_get_key(cfg, "tokenization") is not None: tokenizer = Trainer.build_tokenizer(cfg) model = Trainer.build_model(cfg) Checkpointer(model, save_dir=cfg.train.output_dir).resume_or_load( cfg.train.load_weight, resume=args.resume ) if try_get_key(cfg, "train.graph.enabled", default=False): model = Trainer.build_graph(cfg, model, is_train=False) test_loader = Trainer.build_test_loader(cfg, tokenizer) if len(test_loader) == 0: logger.info("No dataset in dataloader.test, please set dataset for dataloader.test") _ = Trainer.test(cfg, test_loader, model) return trainer = Trainer(cfg) return trainer.train() if __name__ == "__main__": args = default_argument_parser().parse_args() main(args)
5,384
35.385135
98
py
libai
libai-main/projects/Stable_Diffusion/configs/lora_config.py
from omegaconf import OmegaConf from libai.config import get_config from libai.config import LazyCall from libai.data.build import build_nlp_train_loader, build_nlp_test_loader from projects.Stable_Diffusion.dataset import DreamBoothDataset from projects.Stable_Diffusion.modeling import StableDiffusion from transformers import CLIPTokenizer optim = get_config("common/optim.py").optim graph = get_config("common/models/graph.py").graph train = get_config("common/train.py").train graph.global_mode.enabled = True dataloader = OmegaConf.create() dataloader.train = LazyCall(build_nlp_train_loader)( dataset=[ LazyCall(DreamBoothDataset)( instance_data_root="/path/to/demo_dog/", instance_prompt="a photo of sks dog", tokenizer=CLIPTokenizer, tokenizer_pretrained_folder=["CompVis/stable-diffusion-v1-4", "tokenizer"], ) ], num_workers=4, ) optim.lr = 5e-4 model = LazyCall(StableDiffusion)(model_path="CompVis/stable-diffusion-v1-4", train_with_lora=True) train.update( dict( rdma_enabled=True, activation_checkpoint=dict(enabled=True), zero_optimization=dict( enabled=True, stage=2, ), checkpointer=dict(period=5000000), amp=dict(enabled=True), output_dir="output/stable_diffusion/", train_micro_batch_size=1, test_micro_batch_size=1, train_epoch=0, train_iter=800, log_period=1, warmup_ratio=0.01, dist=dict( data_parallel_size=1, tensor_parallel_size=1, pipeline_parallel_size=1, pipeline_stage_id=None, pipeline_num_layers=None, ), evaluation=dict( enabled=False, ), ) )
1,799
28.032258
99
py
libai
libai-main/projects/Stable_Diffusion/configs/prior_preservation_config.py
from omegaconf import OmegaConf import oneflow as flow from libai.optim import get_default_optimizer_params from libai.config import LazyCall from libai.data.build import build_nlp_train_loader, build_nlp_test_loader from projects.Stable_Diffusion.dataset import DreamBoothDataset from transformers import CLIPTokenizer from .dreambooth_config import ( train, optim, graph, model, ) optim.lr = 2e-6 model.train_text_encoder = True dataloader = OmegaConf.create() dataloader.train = LazyCall(build_nlp_train_loader)( dataset=[ LazyCall(DreamBoothDataset)( instance_data_root="path/to/demo_dog/", instance_prompt="a photo of sks dog", class_data_root="/path/to/prior_dog/", class_prompt="a photo of dog", tokenizer=CLIPTokenizer, tokenizer_pretrained_folder=["CompVis/stable-diffusion-v1-4", "tokenizer"], ) ], num_workers=4, ) train.train_iter = 2000 train.log_period = 10
993
25.864865
87
py
libai
libai-main/projects/Stable_Diffusion/configs/dreambooth_config.py
from omegaconf import OmegaConf from libai.config import get_config from libai.config import LazyCall from libai.data.build import build_nlp_train_loader, build_nlp_test_loader from projects.Stable_Diffusion.dataset import DreamBoothDataset from projects.Stable_Diffusion.modeling import StableDiffusion from transformers import CLIPTokenizer optim = get_config("common/optim.py").optim graph = get_config("common/models/graph.py").graph train = get_config("common/train.py").train graph.global_mode.enabled = True dataloader = OmegaConf.create() dataloader.train = LazyCall(build_nlp_train_loader)( dataset=[ LazyCall(DreamBoothDataset)( instance_data_root="/path/to/demo_dog/", instance_prompt="a photo of sks dog", tokenizer=CLIPTokenizer, tokenizer_pretrained_folder=["CompVis/stable-diffusion-v1-4", "tokenizer"], ) ], num_workers=4, ) optim.lr = 5e-6 model = LazyCall(StableDiffusion)(model_path="CompVis/stable-diffusion-v1-4") train.update( dict( rdma_enabled=True, activation_checkpoint=dict(enabled=True), zero_optimization=dict( enabled=True, stage=2, ), checkpointer=dict(period=5000000), amp=dict(enabled=True), output_dir="output/stable_diffusion/", train_micro_batch_size=1, test_micro_batch_size=1, train_epoch=0, train_iter=800, log_period=1, warmup_ratio=0.01, dist=dict( data_parallel_size=1, tensor_parallel_size=1, pipeline_parallel_size=1, pipeline_stage_id=None, pipeline_num_layers=None, ), evaluation=dict( enabled=False, ), ) )
1,777
27.677419
87
py
libai
libai-main/projects/Stable_Diffusion/configs/config.py
from omegaconf import OmegaConf from libai.config import get_config from libai.config import LazyCall from libai.data.build import build_nlp_train_loader, build_nlp_test_loader from projects.Stable_Diffusion.dataset import TXTDataset from projects.Stable_Diffusion.modeling import StableDiffusion from transformers import CLIPTokenizer optim = get_config("common/optim.py").optim graph = get_config("common/models/graph.py").graph train = get_config("common/train.py").train graph.global_mode.enabled = True dataloader = OmegaConf.create() dataloader.train = LazyCall(build_nlp_train_loader)( dataset=[ LazyCall(TXTDataset)( foloder_name="/path/to/mscoco/00000", tokenizer=CLIPTokenizer, tokenizer_pretrained_folder=["CompVis/stable-diffusion-v1-4", "tokenizer"], ) ], num_workers=4, ) model = LazyCall(StableDiffusion)(model_path="CompVis/stable-diffusion-v1-4") train.update( dict( rdma_enabled=True, activation_checkpoint=dict(enabled=True), zero_optimization=dict( enabled=True, stage=2, ), checkpointer=dict(period=5000000), amp=dict(enabled=True), output_dir="output/stable_diffusion/", train_micro_batch_size=1, test_micro_batch_size=1, train_epoch=0, train_iter=20, log_period=1, warmup_ratio=0.01, dist=dict( data_parallel_size=1, tensor_parallel_size=1, pipeline_parallel_size=1, pipeline_stage_id=None, pipeline_num_layers=None, ), evaluation=dict( enabled=False, ), ) )
1,693
27.233333
87
py
libai
libai-main/tests/test_trainer.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import sys import oneflow as flow from omegaconf import OmegaConf from oneflow.utils.data import DataLoader, TensorDataset sys.path.append(".") from libai.config import LazyCall, default_argument_parser from libai.engine import DefaultTrainer, default_setup from libai.optim import get_default_optimizer_params from libai.scheduler import WarmupMultiStepLR from tests.layers.test_trainer_model import build_graph, build_model def setup(args): """ Create configs and perform basic setups. """ cfg = OmegaConf.create() cfg.train = dict( output_dir="./demo_output", train_micro_batch_size=32, test_micro_batch_size=32, dist=dict( data_parallel_size=1, tensor_parallel_size=1, pipeline_parallel_size=1, pipeline_num_layers=4, ), start_iter=0, train_iter=20, train_epoch=1, warmup_ratio=0.05, lr_warmup_fraction=0.01, lr_decay_iter=6000, eval_period=1000, log_period=1, checkpointer=dict(period=100), nccl_fusion_threshold_mb=16, nccl_fusion_max_ops=24, scheduler=LazyCall(WarmupMultiStepLR)( warmup_factor=0.001, # alpha=0.01, warmup_method="linear", milestones=[0.1, 0.2], ), ) cfg.optim = LazyCall(flow.optim.AdamW)( parameters=LazyCall(get_default_optimizer_params)( # parameters.model is meant to be set to the model object, before # instantiating the optimizer. clip_grad_max_norm=1.0, clip_grad_norm_type=2.0, weight_decay_norm=0.0, weight_decay_bias=0.0, ), lr=1e-4, weight_decay=0.01, betas=(0.9, 0.999), do_bias_correction=True, ) cfg.graph = dict( enabled=True, ) default_setup(cfg, args) return cfg class DemoTrainer(DefaultTrainer): @classmethod def build_model(cls, cfg): """ Returns: flow.nn.Module: It now calls :func:`libai.layers.build_model`. Overwrite it if you'd like a different model. """ model = build_model(cfg) return model @classmethod def build_graph(cls, cfg, model, optimizer=None, lr_scheduler=None, is_train=True): return build_graph(cfg, model, optimizer, lr_scheduler) @classmethod def get_batch(cls, data): return [ flow.randn( 32, 512, sbp=flow.sbp.split(0), placement=flow.placement("cuda", [0]), ) ] @classmethod def build_train_loader(cls, cfg, tokenizer=None): return ( DataLoader( TensorDataset(flow.randn(1000)), batch_size=cfg.train.train_micro_batch_size ), None, None, ) @classmethod def build_test_loader(cls, cfg): return [] def main(args): cfg = setup(args) trainer = DemoTrainer(cfg) # trainer.resume_or_load(resume=args.resume) return trainer.train() if __name__ == "__main__": args = default_argument_parser().parse_args() main(args)
3,870
26.06993
92
py
libai
libai-main/tests/test_file_io.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """ Unittests followed https://github.com/facebookresearch/iopath/blob/v0.1.8/tests/test_file_io.py """ import os import shutil import tempfile import unittest import uuid from typing import Optional from unittest.mock import MagicMock from libai.utils.file_io import LazyPath, PathManagerBase, PathManagerFactory, g_pathmgr class TestNativeIO(unittest.TestCase): _tmpdir: Optional[str] = None _filename: Optional[str] = None _tmpfile: Optional[str] = None _tmpfile_contents = "Hello, World" _pathmgr = PathManagerBase() @classmethod def setUpClass(cls) -> None: cls._tmpdir = tempfile.mkdtemp() cls._filename = "test.txt" # pyre-ignore with open(os.path.join(cls._tmpdir, cls._filename), "w") as f: cls._tmpfile = f.name f.write(cls._tmpfile_contents) f.flush() @classmethod def tearDownClass(cls) -> None: # Cleanup temp working dir. if cls._tmpdir is not None: shutil.rmtree(cls._tmpdir) # type: ignore def setUp(self) -> None: # Reset class variables set by methods before each test. self._pathmgr.set_cwd(None) self._pathmgr._native_path_handler._non_blocking_io_manager = None self._pathmgr._native_path_handler._non_blocking_io_executor = None self._pathmgr._async_handlers.clear() def test_open(self) -> None: # pyre-ignore with self._pathmgr.open(self._tmpfile, "r") as f: self.assertEqual(f.read(), self._tmpfile_contents) def test_factory_open(self) -> None: with g_pathmgr.open(self._tmpfile, "r") as f: self.assertEqual(f.read(), self._tmpfile_contents) _pathmgr = PathManagerFactory.get("test_pm") with _pathmgr.open(self._tmpfile, "r") as f: self.assertEqual(f.read(), self._tmpfile_contents) PathManagerFactory.remove("test_pm") def test_open_args(self) -> None: self._pathmgr.set_strict_kwargs_checking(True) f = self._pathmgr.open( self._tmpfile, # type: ignore mode="r", buffering=1, encoding="UTF-8", errors="ignore", newline=None, closefd=True, opener=None, ) f.close() def test_get_local_path(self) -> None: self.assertEqual( # pyre-ignore self._pathmgr.get_local_path(self._tmpfile), self._tmpfile, ) def test_get_local_path_forced(self) -> None: self.assertEqual( # pyre-ignore self._pathmgr.get_local_path(self._tmpfile, force=True), self._tmpfile, ) def test_exists(self) -> None: # pyre-ignore self.assertTrue(self._pathmgr.exists(self._tmpfile)) # pyre-ignore fake_path = os.path.join(self._tmpdir, uuid.uuid4().hex) self.assertFalse(self._pathmgr.exists(fake_path)) def test_isfile(self) -> None: self.assertTrue(self._pathmgr.isfile(self._tmpfile)) # pyre-ignore # This is a directory, not a file, so it should fail self.assertFalse(self._pathmgr.isfile(self._tmpdir)) # pyre-ignore # This is a non-existing path, so it should fail fake_path = os.path.join(self._tmpdir, uuid.uuid4().hex) # pyre-ignore self.assertFalse(self._pathmgr.isfile(fake_path)) def test_isdir(self) -> None: # pyre-ignore self.assertTrue(self._pathmgr.isdir(self._tmpdir)) # This is a file, not a directory, so it should fail # pyre-ignore self.assertFalse(self._pathmgr.isdir(self._tmpfile)) # This is a non-existing path, so it should fail # pyre-ignore fake_path = os.path.join(self._tmpdir, uuid.uuid4().hex) self.assertFalse(self._pathmgr.isdir(fake_path)) def test_ls(self) -> None: # Create some files in the tempdir to ls out. root_dir = os.path.join(self._tmpdir, "ls") # pyre-ignore os.makedirs(root_dir, exist_ok=True) files = sorted(["foo.txt", "bar.txt", "baz.txt"]) for f in files: open(os.path.join(root_dir, f), "a").close() children = sorted(self._pathmgr.ls(root_dir)) self.assertListEqual(children, files) # Cleanup the tempdir shutil.rmtree(root_dir) def test_mkdirs(self) -> None: # pyre-ignore new_dir_path = os.path.join(self._tmpdir, "new", "tmp", "dir") self.assertFalse(self._pathmgr.exists(new_dir_path)) self._pathmgr.mkdirs(new_dir_path) self.assertTrue(self._pathmgr.exists(new_dir_path)) def test_copy(self) -> None: _tmpfile_2 = self._tmpfile + "2" # pyre-ignore _tmpfile_2_contents = "something else" with open(_tmpfile_2, "w") as f: f.write(_tmpfile_2_contents) f.flush() self.assertTrue(self._pathmgr.copy(self._tmpfile, _tmpfile_2, overwrite=True)) with self._pathmgr.open(_tmpfile_2, "r") as f: self.assertEqual(f.read(), self._tmpfile_contents) def test_move(self) -> None: _tmpfile_2 = self._tmpfile + "2" + uuid.uuid4().hex # pyre-ignore _tmpfile_3 = self._tmpfile + "3_" + uuid.uuid4().hex # pyre-ignore _tmpfile_2_contents = "Hello Move" with open(_tmpfile_2, "w") as f: f.write(_tmpfile_2_contents) f.flush() # pyre-ignore self.assertTrue(self._pathmgr.mv(_tmpfile_2, _tmpfile_3)) with self._pathmgr.open(_tmpfile_3, "r") as f: self.assertEqual(f.read(), _tmpfile_2_contents) self.assertFalse(self._pathmgr.exists(_tmpfile_2)) self._pathmgr.rm(_tmpfile_3) def test_symlink(self) -> None: _symlink = self._tmpfile + "_symlink" # pyre-ignore self.assertTrue(self._pathmgr.symlink(self._tmpfile, _symlink)) # pyre-ignore with self._pathmgr.open(_symlink) as f: self.assertEqual(f.read(), self._tmpfile_contents) self.assertEqual(os.readlink(_symlink), self._tmpfile) os.remove(_symlink) def test_rm(self) -> None: # pyre-ignore with open(os.path.join(self._tmpdir, "test_rm.txt"), "w") as f: rm_file = f.name f.write(self._tmpfile_contents) f.flush() self.assertTrue(self._pathmgr.exists(rm_file)) self.assertTrue(self._pathmgr.isfile(rm_file)) self._pathmgr.rm(rm_file) self.assertFalse(self._pathmgr.exists(rm_file)) self.assertFalse(self._pathmgr.isfile(rm_file)) def test_set_cwd(self) -> None: # File not found since cwd not set yet. self.assertFalse(self._pathmgr.isfile(self._filename)) self.assertTrue(self._pathmgr.isfile(self._tmpfile)) # Once cwd is set, relative file path works. self._pathmgr.set_cwd(self._tmpdir) self.assertTrue(self._pathmgr.isfile(self._filename)) # Set cwd to None self._pathmgr.set_cwd(None) self.assertFalse(self._pathmgr.isfile(self._filename)) self.assertTrue(self._pathmgr.isfile(self._tmpfile)) # Set cwd to invalid path with self.assertRaises(ValueError): self._pathmgr.set_cwd("/nonexistent/path") def test_get_path_with_cwd(self) -> None: self._pathmgr.set_cwd(self._tmpdir) # Make sure _get_path_with_cwd() returns correctly. self.assertEqual( self._pathmgr._native_path_handler._get_path_with_cwd(self._filename), self._tmpfile, ) self.assertEqual( self._pathmgr._native_path_handler._get_path_with_cwd("/abs.txt"), "/abs.txt", ) def test_bad_args(self) -> None: # TODO (T58240718): Replace with dynamic checks with self.assertRaises(ValueError): self._pathmgr.copy(self._tmpfile, self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.exists(self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.get_local_path(self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.isdir(self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.isfile(self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.ls(self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.mkdirs(self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.open(self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.opena(self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.rm(self._tmpfile, foo="foo") # type: ignore with self.assertRaises(ValueError): self._pathmgr.set_cwd(self._tmpdir, foo="foo") # type: ignore self._pathmgr.set_strict_kwargs_checking(False) self._pathmgr.copy(self._tmpfile, self._tmpfile + "2", foo="foo") # type: ignore self._pathmgr.exists(self._tmpfile, foo="foo") # type: ignore self._pathmgr.get_local_path(self._tmpfile, foo="foo") # type: ignore self._pathmgr.isdir(self._tmpfile, foo="foo") # type: ignore self._pathmgr.isfile(self._tmpfile, foo="foo") # type: ignore self._pathmgr.ls(self._tmpdir, foo="foo") # type: ignore self._pathmgr.mkdirs(self._tmpdir, foo="foo") # type: ignore f = self._pathmgr.open(self._tmpfile, foo="foo") # type: ignore f.close() # pyre-ignore with open(os.path.join(self._tmpdir, "test_rm.txt"), "w") as f: rm_file = f.name f.write(self._tmpfile_contents) f.flush() self._pathmgr.rm(rm_file, foo="foo") # type: ignore class TestLazyPath(unittest.TestCase): _pathmgr = PathManagerBase() def test_materialize(self) -> None: f = MagicMock(return_value="test") x = LazyPath(f) f.assert_not_called() p = os.fspath(x) f.assert_called() self.assertEqual(p, "test") p = os.fspath(x) # should only be called once f.assert_called_once() self.assertEqual(p, "test") def test_join(self) -> None: f = MagicMock(return_value="test") x = LazyPath(f) p = os.path.join(x, "a.txt") f.assert_called_once() self.assertEqual(p, "test/a.txt") def test_getattr(self) -> None: x = LazyPath(lambda: "abc") with self.assertRaises(AttributeError): x.startswith("ab") _ = os.fspath(x) self.assertTrue(x.startswith("ab")) def test_PathManager(self) -> None: x = LazyPath(lambda: "./") output = self._pathmgr.ls(x) # pyre-ignore output_gt = self._pathmgr.ls("./") self.assertEqual(sorted(output), sorted(output_gt)) def test_getitem(self) -> None: x = LazyPath(lambda: "abc") with self.assertRaises(TypeError): x[0] _ = os.fspath(x) self.assertEqual(x[0], "a") if __name__ == "__main__": unittest.main()
12,018
37.155556
95
py
libai
libai-main/tests/__init__.py
0
0
0
py
libai
libai-main/tests/test_scheduler.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import math import unittest from unittest import TestCase import numpy as np import oneflow as flow import oneflow.nn as nn from libai.scheduler import ( WarmupCosineLR, WarmupExponentialLR, WarmupMultiStepLR, WarmupPolynomialLR, WarmupStepLR, ) # @unittest.skip("Bugs in warmup scheduler") class TestScheduler(TestCase): def test_warmup_multistep(self): p = nn.Parameter(flow.zeros(0)) opt = flow.optim.SGD([p], lr=5.0) sched = WarmupMultiStepLR( optimizer=opt, max_iter=10, milestones=[10, 15, 20], gamma=0.1, warmup_factor=0.001, warmup_iter=5, warmup_method="linear", ) p.sum().backward() opt.step() lrs = [0.005] for _ in range(30): sched.step() lrs.append(opt.param_groups[0]["lr"]) self.assertTrue(np.allclose(lrs[:5], [0.005, 1.004, 2.003, 3.002, 4.001])) self.assertTrue(np.allclose(lrs[5:10], 5.0)) self.assertTrue(np.allclose(lrs[10:15], 0.5)) self.assertTrue(np.allclose(lrs[15:20], 0.05)) self.assertTrue(np.allclose(lrs[20:], 0.005)) def test_warmup_step(self): p = nn.Parameter(flow.zeros(0)) opt = flow.optim.SGD([p], lr=5.0) sched = WarmupStepLR( optimizer=opt, max_iter=10, step_size=10, gamma=0.1, warmup_factor=0.001, warmup_iter=5, warmup_method="linear", ) p.sum().backward() opt.step() lrs = [0.005] for _ in range(30): sched.step() lrs.append(opt.param_groups[0]["lr"]) self.assertTrue(np.allclose(lrs[:5], [0.005, 1.004, 2.003, 3.002, 4.001])) self.assertTrue(np.allclose(lrs[5:10], 5.0)) self.assertTrue(np.allclose(lrs[10:20], 0.5)) self.assertTrue(np.allclose(lrs[20:30], 0.05)) self.assertTrue(np.allclose(lrs[30:], 0.005)) def test_warmup_cosine(self): p = nn.Parameter(flow.zeros(0)) opt = flow.optim.SGD([p], lr=5.0) sched = WarmupCosineLR( optimizer=opt, max_iter=30, warmup_factor=0.001, warmup_iter=5, warmup_method="linear", ) p.sum().backward() opt.step() self.assertEqual(opt.param_groups[0]["lr"], 0.005) lrs = [0.005] for _ in range(30): sched.step() lrs.append(opt.param_groups[0]["lr"]) for idx, lr in enumerate(lrs): expected_cosine = 2.5 * (1.0 + math.cos(math.pi * idx / 30)) if idx >= 5: self.assertAlmostEqual(lr, expected_cosine) else: self.assertNotAlmostEqual(lr, expected_cosine) def test_warmup_exponential(self): p = nn.Parameter(flow.zeros(0)) opt = flow.optim.SGD([p], lr=5.0) sched = WarmupExponentialLR( optimizer=opt, max_iter=10, gamma=0.1, warmup_factor=0.001, warmup_iter=5, warmup_method="linear", ) p.sum().backward() opt.step() self.assertEqual(opt.param_groups[0]["lr"], 0.005) lrs = [0.005] def _get_exponential_lr(base_lr, gamma, max_iters, warmup_iters): valid_values = [] for idx in range(warmup_iters, max_iters + 1): valid_values.append(base_lr * (gamma ** idx)) return valid_values for _ in range(30): sched.step() lrs.append(opt.param_groups[0]["lr"]) self.assertTrue( np.allclose( lrs[:5], [0.005, 0.00401, 0.0030199999999999997, 0.00203, 0.0010399999999999997] ) ) valid_intermediate_values = _get_exponential_lr( base_lr=5.0, gamma=0.1, max_iters=30, warmup_iters=5 ) self.assertEqual(lrs[5:], valid_intermediate_values) def test_warmup_polynomial(self): p = nn.Parameter(flow.zeros(0)) opt = flow.optim.SGD([p], lr=5.0) sched = WarmupPolynomialLR( optimizer=opt, max_iter=30, warmup_factor=0.001, warmup_iter=0, end_learning_rate=1e-4, power=1.0, cycle=False, warmup_method="linear", ) # self.assertEqual(opt.param_groups[0]["lr"], 0.005) # lrs = [0.005] lrs = [5.0] # lr_scheduler first invoke result def _get_polynomial_lr( base_lr, max_iters, warmup_iters, end_lr=1e-4, power=1.0, cycle=False ): valid_values = [] decay_steps = max_iters - warmup_iters for step in range(max_iters - warmup_iters): if cycle: if step == 0: step = 1 decay_steps = decay_steps * math.ceil(step / decay_steps) else: step = min(step, decay_steps) valid_values.append( (base_lr - end_lr) * ((1 - step / decay_steps) ** power) + end_lr ) return valid_values for _ in range(29): sched.step() # only invoke (max_iter-1), because the first invoke is done when init lrs.append(opt.param_groups[0]["lr"]) # self.assertTrue(np.allclose(lrs[:5], [0.005, 1.004, 2.003, 3.002, 4.001])) valid_intermediate_values = _get_polynomial_lr(base_lr=5.0, max_iters=30, warmup_iters=0) # self.assertEqual(lrs[5:30], valid_intermediate_values) self.assertEqual(lrs, valid_intermediate_values) if __name__ == "__main__": unittest.main()
6,402
31.015
97
py
libai
libai-main/tests/test_optim.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # Copyright (c) Facebook, Inc. and its affiliates. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """ Unittests followed https://github.com/facebookresearch/detectron2/blob/main/tests/test_solver.py """ import unittest from libai.optim.build import _expand_param_groups, reduce_param_groups class TestOptimizer(unittest.TestCase): def testExpandParamGroups(self): params = [ {"params": ["p1", "p2", "p3", "p4"], "lr": 1.0, "weight_decay": 3.0}, {"params": ["p2", "p3", "p5"], "lr": 2.0, "momentum": 2.0}, {"params": ["p1"], "weight_decay": 4.0}, ] out = _expand_param_groups(params) gt = [ dict(params=["p1"], lr=1.0, weight_decay=4.0), # noqa dict(params=["p2"], lr=2.0, weight_decay=3.0, momentum=2.0), # noqa dict(params=["p3"], lr=2.0, weight_decay=3.0, momentum=2.0), # noqa dict(params=["p4"], lr=1.0, weight_decay=3.0), # noqa dict(params=["p5"], lr=2.0, momentum=2.0), # noqa ] self.assertEqual(out, gt) def testReduceParamGroups(self): params = [ dict(params=["p1"], lr=1.0, weight_decay=4.0), # noqa dict(params=["p2", "p6"], lr=2.0, weight_decay=3.0, momentum=2.0), # noqa dict(params=["p3"], lr=2.0, weight_decay=3.0, momentum=2.0), # noqa dict(params=["p4"], lr=1.0, weight_decay=3.0), # noqa dict(params=["p5"], lr=2.0, momentum=2.0), # noqa ] gt_groups = [ {"lr": 1.0, "weight_decay": 4.0, "params": ["p1"]}, {"lr": 2.0, "weight_decay": 3.0, "momentum": 2.0, "params": ["p2", "p6", "p3"]}, {"lr": 1.0, "weight_decay": 3.0, "params": ["p4"]}, {"lr": 2.0, "momentum": 2.0, "params": ["p5"]}, ] out = reduce_param_groups(params) self.assertTrue(out, gt_groups) if __name__ == "__main__": unittest.main()
2,519
39
96
py
libai
libai-main/tests/fixtures/utils.py
from libai.utils.download import download fixtrue_urls = { "sample_text.txt": "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/fixtures/sample_text.txt", # noqa "spiece.model": "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/fixtures/spiece.model", # noqa "test_sentencepiece.model": "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/fixtures/test_sentencepiece.model", # noqa } BASE_DIR = "tests/fixtures" def get_fixtures(fixture_dir): fixture_name = fixture_dir.split("/")[-1] if fixture_name not in fixtrue_urls: raise RuntimeError("{} not available in LiBai tests fixtrues!".format(fixture_name)) return download(fixtrue_urls[fixture_name], BASE_DIR)
773
44.529412
150
py
libai
libai-main/tests/config/test_lazy_config.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # Copyright (c) Facebook, Inc. and its affiliates. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """ Unittests followed https://github.com/facebookresearch/detectron2/blob/main/tests/config/test_lazy_config.py """ import os import unittest import tempfile from itertools import count from libai.config import LazyConfig, LazyCall from omegaconf import DictConfig class TestLazyPythonConfig(unittest.TestCase): def setUp(self): self.root_filename = os.path.join(os.path.dirname(__file__), "root_cfg.py") def test_load(self): cfg = LazyConfig.load(self.root_filename) self.assertEqual(cfg.dir1a_dict.a, "modified") self.assertEqual(cfg.dir1b_dict.a, 1) self.assertEqual(cfg.lazyobj.x, "base_a_1") cfg.lazyobj.x = "new_x" # reload cfg = LazyConfig.load(self.root_filename) self.assertEqual(cfg.lazyobj.x, "base_a_1") def test_save_load(self): cfg = LazyConfig.load(self.root_filename) with tempfile.TemporaryDirectory(prefix="detectron2") as d: fname = os.path.join(d, "test_config.yaml") LazyConfig.save(cfg, fname) cfg2 = LazyConfig.load(fname) self.assertEqual(cfg2.lazyobj._target_, "itertools.count") self.assertEqual(cfg.lazyobj._target_, count) cfg2.lazyobj.pop("_target_") cfg.lazyobj.pop("_target_") # the rest are equal self.assertEqual(cfg, cfg2) def test_failed_save(self): cfg = DictConfig({"x": lambda: 3}, flags={"allow_objects": True}) with tempfile.TemporaryDirectory(prefix="detectron2") as d: fname = os.path.join(d, "test_config.yaml") LazyConfig.save(cfg, fname) self.assertTrue(os.path.exists(fname)) self.assertTrue(os.path.exists(fname + ".pkl")) def test_overrides(self): cfg = LazyConfig.load(self.root_filename) LazyConfig.apply_overrides(cfg, ["lazyobj.x=123", 'dir1b_dict.a="123"']) self.assertEqual(cfg.dir1b_dict.a, "123") self.assertEqual(cfg.lazyobj.x, 123) def test_invalid_overrides(self): cfg = LazyConfig.load(self.root_filename) with self.assertRaises(KeyError): LazyConfig.apply_overrides(cfg, ["lazyobj.x.xxx=123"]) def test_to_py(self): cfg = LazyConfig.load(self.root_filename) cfg.lazyobj.x = { "a": 1, "b": 2, "c": LazyCall(count)(x={"r": "a", "s": 2.4, "t": [1, 2, 3, "z"]}), } cfg.list = ["a", 1, "b", 3.2] py_str = LazyConfig.to_py(cfg) expected = """cfg.dir1a_dict.a = "modified" cfg.dir1a_dict.b = 2 cfg.dir1b_dict.a = 1 cfg.dir1b_dict.b = 2 cfg.lazyobj = itertools.count( x={ "a": 1, "b": 2, "c": itertools.count(x={"r": "a", "s": 2.4, "t": [1, 2, 3, "z"]}), }, y="base_a_1_from_b", ) cfg.list = ["a", 1, "b", 3.2] """ self.assertEqual(py_str, expected)
3,540
33.378641
89
py
libai
libai-main/tests/config/test_instantiate_config.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # Copyright (c) Facebook, Inc. and its affiliates. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """ Unittests followed https://github.com/facebookresearch/detectron2/blob/main/tests/config/test_instantiate_config.py """ from collections import namedtuple import os import unittest import yaml import tempfile from libai.config import instantiate, LazyCall from omegaconf import OmegaConf from dataclasses import dataclass from omegaconf import __version__ as oc_version OC_VERSION = tuple(int(x) for x in oc_version.split(".")[:2]) class ShapeSpec(namedtuple("_ShapeSpec", ["channels", "width"])): """ A simple structure that contains basic shape specification about a tensor. It is often used as the auxiliary inputs/outputs of models, to complement the lack of shape inference ability among pytorch modules. Attributes: channels: width: """ def __new__(cls, channels=None, width=None): return super().__new__(cls, channels, width) class TestClass: def __init__(self, int_arg, list_arg=None, dict_arg=None, extra_arg=None) -> None: self.int_arg = int_arg self.list_arg = list_arg self.dict_arg = dict_arg self.extra_arg = extra_arg def __call__(self, call_arg): return call_arg + self.int_arg @dataclass class TestDataClass: x: int y: str @unittest.skipIf(OC_VERSION < (2, 1), "omegaconf version too old") class TestConstruction(unittest.TestCase): def test_basic_construct(self): objconf = LazyCall(TestClass)( int_arg=3, list_arg=[10], dict_arg={}, extra_arg=LazyCall(TestClass)(int_arg=4, list_arg="${..list_arg}"), ) obj = instantiate(objconf) self.assertIsInstance(obj, TestClass) self.assertEqual(obj.int_arg, 3) self.assertEqual(obj.extra_arg.int_arg, 4) self.assertEqual(obj.extra_arg.list_arg, obj.list_arg) objconf.extra_arg.list_arg = [5] obj = instantiate(objconf) self.assertIsInstance(obj, TestClass) self.assertEqual(obj.extra_arg.list_arg, [5]) def test_instantiate_other_obj(self): # do nothing for other obj self.assertEqual(instantiate(5), 5) x = [3, 4, 5] self.assertEqual(instantiate(x), x) x = TestClass(1) self.assertIs(instantiate(x), x) x = {"xx": "yy"} self.assertEqual(instantiate(x), x) def test_instantiate_lazy_target(self): # _target_ is result of instantiate objconf = LazyCall(LazyCall(len)(int_arg=3))(call_arg=4) objconf._target_._target_ = TestClass self.assertEqual(instantiate(objconf), 7) def test_instantiate_lst(self): lst = [1, 2, LazyCall(TestClass)(int_arg=1)] x = LazyCall(TestClass)( int_arg=lst ) # list as an argument should be recursively instantiated x = instantiate(x).int_arg self.assertEqual(x[:2], [1, 2]) self.assertIsInstance(x[2], TestClass) self.assertEqual(x[2].int_arg, 1) def test_instantiate_namedtuple(self): x = LazyCall(TestClass)(int_arg=ShapeSpec(channels=1, width=3)) # test serialization with tempfile.TemporaryDirectory() as d: fname = os.path.join(d, "lb_test.yaml") OmegaConf.save(x, fname) with open(fname) as f: x = yaml.unsafe_load(f) x = instantiate(x) self.assertIsInstance(x.int_arg, ShapeSpec) self.assertEqual(x.int_arg.channels, 1) def test_bad_lazycall(self): with self.assertRaises(Exception): LazyCall(3) def test_instantiate_dataclass(self): a = LazyCall(TestDataClass)(x=1, y="s") a = instantiate(a) self.assertEqual(a.x, 1) self.assertEqual(a.y, "s") def test_instantiate_no_recursive(self): def helper_func(obj): self.assertNotIsInstance(obj, TestClass) obj = instantiate(obj) self.assertIsInstance(obj, TestClass) return obj.int_arg objconf = LazyCall(helper_func)(obj=LazyCall(TestClass)(int_arg=4)) self.assertEqual(instantiate(objconf, _recursive_=False), 4) if __name__ == "__main__": unittest.main()
4,865
31.657718
96
py
libai
libai-main/tests/config/root_cfg.py
# Copyright (c) Facebook, Inc. and its affiliates. from itertools import count from libai.config import LazyCall from .dir1.dir1_a import dir1a_dict, dir1a_str dir1a_dict.a = "modified" # modification above won't affect future imports from .dir1.dir1_b import dir1b_dict, dir1b_str lazyobj = LazyCall(count)(x=dir1a_str, y=dir1b_str)
340
21.733333
51
py
libai
libai-main/tests/config/dir1/dir1_a.py
# Copyright (c) Facebook, Inc. and its affiliates. dir1a_str = "base_a_1" dir1a_dict = {"a": 1, "b": 2}
104
25.25
50
py
libai
libai-main/tests/config/dir1/dir1_b.py
# Copyright (c) Facebook, Inc. and its affiliates. from libai.config import LazyConfig # equivalent to relative import dir1a_str, dir1a_dict = LazyConfig.load_rel("dir1_a.py", ("dir1a_str", "dir1a_dict")) dir1b_str = dir1a_str + "_from_b" dir1b_dict = dir1a_dict # Every import is a reload: not modified by other config files assert dir1a_dict.a == 1
354
28.583333
85
py
libai
libai-main/tests/models/test_vit.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import oneflow as flow import oneflow.unittest from flowvision.loss.cross_entropy import SoftTargetCrossEntropy import libai.utils.distributed as dist from configs.common.models.vit.vit_small_patch16_224 import model from libai.config import LazyCall, LazyConfig from libai.data.datasets import CIFAR10Dataset from libai.engine import DefaultTrainer from libai.engine.default import _check_batch_size from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/cifar10/cifar-10-python.tar.gz" # noqa DATA_MD5 = "c58f30108f718f92721af3b95e74349a" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_vit") setup_logger(distributed_rank=dist.get_rank()) class TestViTModel(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join(os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "vit_data") cfg = LazyConfig.load("configs/vit_imagenet.py") # set model cfg.model = model cfg.model.num_classes = 10 cfg.model.depth = 6 cfg.model.loss_func = LazyCall(SoftTargetCrossEntropy)() # prepare data path if dist.get_local_rank() == 0: get_data_from_cache(DATA_URL, cache_dir, md5=DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() data_path = get_data_from_cache(DATA_URL, cache_dir, md5=DATA_MD5) cfg.dataloader.train.dataset[0]._target_ = CIFAR10Dataset cfg.dataloader.train.dataset[0].root = "/".join(data_path.split("/")[:-1]) cfg.dataloader.train.dataset[0].download = True cfg.dataloader.train.num_workers = 0 cfg.dataloader.test[0].dataset._target_ = CIFAR10Dataset cfg.dataloader.test[0].dataset.train = False cfg.dataloader.test[0].dataset.root = "/".join(data_path.split("/")[:-1]) cfg.dataloader.test[0].dataset.download = True cfg.dataloader.test[0].num_workers = 0 # refine mixup cfg cfg.dataloader.train.mixup_func.num_classes = 10 # set training config cfg.train.train_epoch = 0 cfg.train.train_iter = 10 cfg.train.evaluation.eval_period = 10 cfg.train.evaluation.eval_iter = 10 cfg.train.log_period = 1 cfg.train.train_micro_batch_size = 8 cfg.train.num_accumulation_steps = 1 cfg.train.resume = False cfg.train.output_dir = TEST_OUTPUT cfg.train.activation_checkpoint.enabled = True cfg.train.amp.enabled = True cfg.train.rdma_enabled = False self.cfg = cfg @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_vit_eager_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 # pipeline parallelism not supported in eager global now! self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_vit_eager_with_pipeline_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 1 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 4 self.cfg.train.dist.pipeline_num_layers = self.cfg.model.depth dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_vit_graph_with_data_tensor_parallel(self): self.cfg.train.num_accumulation_steps = 1 # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_vit_graph_with_data_tensor_pipeline_parallel(self): self.cfg.train.num_accumulation_steps = 4 # set distributed config self.cfg.train.dist.data_parallel_size = 2 # change to 2 when 2d sbp bugfix self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 2 self.cfg.train.dist.pipeline_num_layers = self.cfg.model.depth dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_vit_with_zero(self): # set distributed config self.cfg.train.dist.data_parallel_size = 4 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True self.cfg.train.zero_optimization.enabled = True self.cfg.train.zero_optimization.stage = 3 trainer = DefaultTrainer(self.cfg) trainer.train() if __name__ == "__main__": unittest.main()
6,402
34.77095
125
py
libai
libai-main/tests/models/test_swin.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import oneflow as flow import oneflow.unittest from flowvision.loss.cross_entropy import SoftTargetCrossEntropy import libai.utils.distributed as dist from configs.common.models.swin.swin_tiny_patch4_window7_224 import model from libai.config import LazyConfig from libai.data.datasets import CIFAR10Dataset from libai.engine import DefaultTrainer from libai.engine.default import _check_batch_size from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/cifar10/cifar-10-python.tar.gz" # noqa DATA_MD5 = "c58f30108f718f92721af3b95e74349a" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_swin") setup_logger(distributed_rank=dist.get_rank()) class TestSwinModel(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join(os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "swin_data") cfg = LazyConfig.load("configs/swin_cifar100.py") # set model cfg.model = model cfg.model.cfg.num_classes = 10 cfg.model.cfg.loss_func = SoftTargetCrossEntropy() # prepare data path if dist.get_local_rank() == 0: get_data_from_cache(DATA_URL, cache_dir, md5=DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() data_path = get_data_from_cache(DATA_URL, cache_dir, md5=DATA_MD5) cfg.dataloader.train.dataset[0]._target_ = CIFAR10Dataset cfg.dataloader.train.dataset[0].root = "/".join(data_path.split("/")[:-1]) cfg.dataloader.train.dataset[0].download = True cfg.dataloader.train.num_workers = 0 cfg.dataloader.test[0].dataset._target_ = CIFAR10Dataset cfg.dataloader.test[0].dataset.train = False cfg.dataloader.test[0].dataset.root = "/".join(data_path.split("/")[:-1]) cfg.dataloader.test[0].dataset.download = True cfg.dataloader.test[0].num_workers = 0 # refine mixup cfg cfg.dataloader.train.mixup_func.num_classes = 10 # set training config cfg.train.train_epoch = 0 cfg.train.train_iter = 10 cfg.train.evaluation.eval_period = 10 cfg.train.evaluation.eval_iter = 10 cfg.train.log_period = 1 cfg.train.train_micro_batch_size = 8 cfg.train.num_accumulation_steps = 1 cfg.train.resume = False cfg.train.output_dir = TEST_OUTPUT cfg.train.activation_checkpoint.enabled = True cfg.train.amp.enabled = True cfg.train.rdma_enabled = False self.cfg = cfg @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_swin_eager_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 # pipeline parallelism not supported in eager global now! self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_swin_eager_with_pipeline_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 1 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 4 self.cfg.train.dist.pipeline_num_layers = sum(self.cfg.model.cfg.depths) dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_swin_graph_with_data_tensor_parallel(self): self.cfg.train.num_accumulation_steps = 1 # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_swin_graph_with_data_tensor_pipeline_parallel(self): self.cfg.train.num_accumulation_steps = 4 # set distributed config self.cfg.train.dist.data_parallel_size = 2 # change to 2 when 2d sbp bugfix self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 2 self.cfg.train.dist.pipeline_num_layers = sum(self.cfg.model.cfg.depths) dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_swin_with_zero(self): # set distributed config self.cfg.train.dist.data_parallel_size = 4 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True self.cfg.train.zero_optimization.enabled = True self.cfg.train.zero_optimization.stage = 3 trainer = DefaultTrainer(self.cfg) trainer.train() if __name__ == "__main__": unittest.main()
6,399
34.955056
125
py
libai
libai-main/tests/models/test_swinv2.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import oneflow as flow import oneflow.unittest from flowvision.loss.cross_entropy import SoftTargetCrossEntropy import libai.utils.distributed as dist from configs.common.models.swinv2.swinv2_tiny_patch4_window8_256 import model from libai.config import LazyConfig from libai.data.datasets import CIFAR10Dataset from libai.engine import DefaultTrainer from libai.engine.default import _check_batch_size from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/cifar10/cifar-10-python.tar.gz" # noqa DATA_MD5 = "c58f30108f718f92721af3b95e74349a" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_swinv2") setup_logger(distributed_rank=dist.get_rank()) class TestSwinV2Model(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join(os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "swinv2_data") cfg = LazyConfig.load("configs/swinv2_cifar100.py") # set model cfg.model = model cfg.model.cfg.num_classes = 10 cfg.model.cfg.loss_func = SoftTargetCrossEntropy() # prepare data path if dist.get_local_rank() == 0: get_data_from_cache(DATA_URL, cache_dir, md5=DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() data_path = get_data_from_cache(DATA_URL, cache_dir, md5=DATA_MD5) cfg.dataloader.train.dataset[0]._target_ = CIFAR10Dataset cfg.dataloader.train.dataset[0].root = "/".join(data_path.split("/")[:-1]) cfg.dataloader.train.dataset[0].download = True cfg.dataloader.train.num_workers = 0 cfg.dataloader.test[0].dataset._target_ = CIFAR10Dataset cfg.dataloader.test[0].dataset.train = False cfg.dataloader.test[0].dataset.root = "/".join(data_path.split("/")[:-1]) cfg.dataloader.test[0].dataset.download = True cfg.dataloader.test[0].num_workers = 0 # refine mixup cfg cfg.dataloader.train.mixup_func.num_classes = 10 # set training config cfg.train.train_epoch = 0 cfg.train.train_iter = 10 cfg.train.evaluation.eval_period = 10 cfg.train.evaluation.eval_iter = 10 cfg.train.log_period = 1 cfg.train.train_micro_batch_size = 8 cfg.train.num_accumulation_steps = 1 cfg.train.resume = False cfg.train.output_dir = TEST_OUTPUT cfg.train.activation_checkpoint.enabled = True cfg.train.amp.enabled = True cfg.train.rdma_enabled = False self.cfg = cfg @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_swinv2_eager_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 # pipeline parallelism not supported in eager global now! self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_swinv2_eager_with_pipeline_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 1 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 4 self.cfg.train.dist.pipeline_num_layers = sum(self.cfg.model.cfg.depths) dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_swinv2_graph_with_data_tensor_parallel(self): self.cfg.train.num_accumulation_steps = 1 # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_swinv2_graph_with_data_tensor_pipeline_parallel(self): self.cfg.train.num_accumulation_steps = 4 # set distributed config self.cfg.train.dist.data_parallel_size = 2 # change to 2 when 2d sbp bugfix self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 2 self.cfg.train.dist.pipeline_num_layers = sum(self.cfg.model.cfg.depths) dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_swinv2_with_zero(self): # set distributed config self.cfg.train.dist.data_parallel_size = 4 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True self.cfg.train.zero_optimization.enabled = True self.cfg.train.zero_optimization.stage = 3 trainer = DefaultTrainer(self.cfg) trainer.train() if __name__ == "__main__": unittest.main()
6,421
35.078652
125
py
libai
libai-main/tests/models/test_mt5.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import oneflow as flow import oneflow.unittest from libai.config import LazyConfig from libai.engine import DefaultTrainer, hooks from libai.engine.default import _check_batch_size from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger VOCAB_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/bert-base-chinese-vocab.txt" # noqa BIN_DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/loss_compara_content_sentence.bin" # noqa IDX_DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/loss_compara_content_sentence.idx" # noqa VOCAB_MD5 = "3b5b76c4aef48ecf8cb3abaafe960f09" BIN_DATA_MD5 = "b842467bd5ea7e52f7a612ea6b4faecc" IDX_DATA_MD5 = "cf5963b8543f0a7a867361eb980f0372" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_mt5") setup_logger(distributed_rank=dist.get_rank()) class TestMT5Model(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join(os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "bert_data") cfg = LazyConfig.load("projects/MT5/configs/mt5_pretrain.py") # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) get_data_from_cache(IDX_DATA_URL, cache_dir, md5=IDX_DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() vocab_path = get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) data_prefix_path = get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) data_prefix = data_prefix_path[:-4] # set tokenizer and data config cfg.tokenization.tokenizer.vocab_file = vocab_path cfg.dataloader.train.dataset[0].data_prefix = data_prefix cfg.dataloader.train.dataset[0].indexed_dataset.data_prefix = data_prefix # FIXME(RenTianhe): fix dataloader worker bug cfg.dataloader.train.num_workers = 0 # set training config cfg.train.train_epoch = 0 cfg.train.train_iter = 10 cfg.train.evaluation.enabled = True cfg.train.evaluation.eval_period = 10 cfg.train.evaluation.eval_iter = 10 cfg.train.log_period = 1 cfg.train.train_micro_batch_size = 8 cfg.train.num_accumulation_steps = 1 cfg.train.resume = False cfg.train.output_dir = TEST_OUTPUT # set model cfg.model.cfg.num_attention_heads = 8 cfg.model.cfg.hidden_size = 384 cfg.model.cfg.hidden_layers = 3 cfg.train.activation_checkpoint.enabled = True cfg.train.amp.enabled = True cfg.train.rdma_enabled = False self.cfg = cfg def build_hooks(self): ret = [ hooks.IterationTimer(), hooks.LRScheduler(), ] if dist.is_main_process(): # run writers in the end, so that evaluation metrics are written ret.append(hooks.PeriodicWriter(self.build_writers(), self.cfg.train.log_period)) return ret DefaultTrainer.build_hooks = build_hooks @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_mt5_eager_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 # pipeline parallelism not supported in eager global now! self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_mt5_eager_with_pipeline_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 1 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 4 self.cfg.train.dist.pipeline_num_layers = 2 * self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_mt5_graph_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_mt5_graph_with_data_tensor_pipeline_parallel(self): self.cfg.train.num_accumulation_steps = 4 # set distributed config self.cfg.train.dist.data_parallel_size = 2 # change to 2 when 2d sbp bugfix self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 2 # encoder_layers + decoder_layers self.cfg.train.dist.pipeline_num_layers = 2 * self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_mt5_with_zero(self): # set distributed config self.cfg.train.dist.data_parallel_size = 4 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True self.cfg.train.zero_optimization.enabled = True self.cfg.train.zero_optimization.stage = 3 trainer = DefaultTrainer(self.cfg) trainer.train() if __name__ == "__main__": unittest.main()
7,176
36.186528
145
py
libai
libai-main/tests/models/test_bert.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import oneflow as flow import oneflow.unittest from libai.config import LazyConfig from libai.engine import DefaultTrainer from libai.engine.default import _check_batch_size from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger VOCAB_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/bert-base-chinese-vocab.txt" # noqa BIN_DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/loss_compara_content_sentence.bin" # noqa IDX_DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/loss_compara_content_sentence.idx" # noqa VOCAB_MD5 = "3b5b76c4aef48ecf8cb3abaafe960f09" BIN_DATA_MD5 = "b842467bd5ea7e52f7a612ea6b4faecc" IDX_DATA_MD5 = "cf5963b8543f0a7a867361eb980f0372" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_bert") setup_logger(distributed_rank=dist.get_rank()) class TestBertModel(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join(os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "bert_data") cfg = LazyConfig.load("configs/bert_large_pretrain.py") # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) get_data_from_cache(IDX_DATA_URL, cache_dir, md5=IDX_DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() vocab_path = get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) data_prefix_path = get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) data_prefix = data_prefix_path[:-4] # set tokenizer and data config cfg.tokenization.tokenizer.vocab_file = vocab_path cfg.dataloader.train.dataset[0].data_prefix = data_prefix cfg.dataloader.train.dataset[0].indexed_dataset.data_prefix = data_prefix # FIXME(RenTianhe): fix dataloader worker bug cfg.dataloader.train.num_workers = 0 # set training config cfg.train.train_epoch = 0 cfg.train.train_iter = 10 cfg.train.evaluation.eval_period = 10 cfg.train.evaluation.eval_iter = 10 cfg.train.log_period = 1 cfg.train.train_micro_batch_size = 8 cfg.train.test_micro_batch_size = 4 cfg.train.num_accumulation_steps = 1 cfg.train.resume = False cfg.train.output_dir = TEST_OUTPUT # set model cfg.model.cfg.num_attention_heads = 8 cfg.model.cfg.hidden_size = 384 cfg.model.cfg.hidden_layers = 4 cfg.train.activation_checkpoint.enabled = True cfg.train.amp.enabled = True cfg.train.rdma_enabled = False self.cfg = cfg @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_bert_eager_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 # pipeline parallelism not supported in eager global now! self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_bert_eager_with_pipeline_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 1 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 4 self.cfg.train.dist.pipeline_num_layers = self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_bert_graph_with_data_tensor_parallel(self): self.cfg.train.num_accumulation_steps = 1 # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_bert_graph_with_data_tensor_pipeline_parallel(self): self.cfg.train.num_accumulation_steps = 4 # set distributed config self.cfg.train.dist.data_parallel_size = 2 # change to 2 when 2d sbp bugfix self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 2 self.cfg.train.dist.pipeline_num_layers = self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_bert_with_zero(self): # set distributed config self.cfg.train.dist.data_parallel_size = 4 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True self.cfg.train.zero_optimization.enabled = True self.cfg.train.zero_optimization.stage = 3 trainer = DefaultTrainer(self.cfg) trainer.train() if __name__ == "__main__": unittest.main()
6,735
36.422222
145
py
libai
libai-main/tests/models/test_roberta.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import oneflow as flow import oneflow.unittest from libai.config import LazyConfig from libai.engine import DefaultTrainer from libai.engine.default import _check_batch_size from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger VOCAB_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/roberta_dataset/roberta-vocab.json" # noqa MERGE_FILE_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/roberta_dataset/roberta-merges.txt" # noqa BIN_DATA_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/roberta_dataset/loss_compara_content_sentence.bin" # noqa IDX_DATA_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/roberta_dataset/loss_compara_content_sentence.idx" # noqa VOCAB_MD5 = "be4d3c6f3f5495426b2c03b334334354" MERGES_FILE_MD5 = "75a37753dd7a28a2c5df80c28bf06e4e" BIN_DATA_MD5 = "b842467bd5ea7e52f7a612ea6b4faecc" IDX_DATA_MD5 = "cf5963b8543f0a7a867361eb980f0372" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_roberta") setup_logger(distributed_rank=dist.get_rank()) class TestRoBERTaModel(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join(os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "roberta_data") cfg = LazyConfig.load("configs/roberta_pretrain.py") # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) get_data_from_cache(MERGE_FILE_URL, cache_dir, md5=MERGES_FILE_MD5) get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) get_data_from_cache(IDX_DATA_URL, cache_dir, md5=IDX_DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() vocab_path = get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) merges_file_path = get_data_from_cache(MERGE_FILE_URL, cache_dir, md5=MERGES_FILE_MD5) data_prefix_path = get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) data_prefix = data_prefix_path[:-4] # set tokenizer and data config cfg.tokenization.tokenizer.vocab_file = vocab_path cfg.tokenization.tokenizer.merges_file = merges_file_path cfg.dataloader.train.dataset[0].data_prefix = data_prefix cfg.dataloader.train.dataset[0].indexed_dataset.data_prefix = data_prefix cfg.dataloader.train.num_workers = 0 # set training config cfg.train.train_epoch = 0 cfg.train.train_iter = 10 cfg.train.evaluation.eval_period = 10 cfg.train.evaluation.eval_iter = 10 cfg.train.log_period = 1 cfg.train.train_micro_batch_size = 8 cfg.train.test_micro_batch_size = 4 cfg.train.num_accumulation_steps = 1 cfg.train.resume = False cfg.train.output_dir = TEST_OUTPUT # set model cfg.model.cfg.num_attention_heads = 8 cfg.model.cfg.hidden_size = 384 cfg.model.cfg.hidden_layers = 4 cfg.train.activation_checkpoint.enabled = True cfg.train.amp.enabled = True cfg.train.rdma_enabled = False self.cfg = cfg @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_roberta_eager_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 # pipeline parallelism not supported in eager global now! self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_roberta_eager_with_pipeline_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 1 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 4 self.cfg.train.dist.pipeline_num_layers = self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_roberta_graph_with_data_tensor_parallel(self): self.cfg.train.num_accumulation_steps = 1 # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_roberta_graph_with_data_tensor_pipeline_parallel(self): self.cfg.train.num_accumulation_steps = 4 # set distributed config self.cfg.train.dist.data_parallel_size = 2 # change to 2 when 2d sbp bugfix self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 2 self.cfg.train.dist.pipeline_num_layers = self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_roberta_with_zero(self): # set distributed config self.cfg.train.dist.data_parallel_size = 4 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True self.cfg.train.zero_optimization.enabled = True self.cfg.train.zero_optimization.stage = 3 trainer = DefaultTrainer(self.cfg) trainer.train() if __name__ == "__main__": unittest.main()
7,129
37.540541
147
py
libai
libai-main/tests/models/__init__.py
0
0
0
py
libai
libai-main/tests/models/test_gpt.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import oneflow as flow import oneflow.unittest from libai.config import LazyConfig from libai.engine import DefaultTrainer, hooks from libai.engine.default import _check_batch_size from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger VOCAB_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/gpt_dataset/gpt2-vocab.json" # noqa MERGE_FILE_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/gpt_dataset/gpt2-merges.txt" # noqa BIN_DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/loss_compara_content_sentence.bin" # noqa IDX_DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/loss_compara_content_sentence.idx" # noqa VOCAB_MD5 = "dffec25a898b1f5e569bec4dffd7e5c0" MERGE_FILE_MD5 = "75a37753dd7a28a2c5df80c28bf06e4e" BIN_DATA_MD5 = "b842467bd5ea7e52f7a612ea6b4faecc" IDX_DATA_MD5 = "cf5963b8543f0a7a867361eb980f0372" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_gpt") setup_logger(distributed_rank=dist.get_rank()) class TestGPTModel(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join(os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "gpt_data") cfg = LazyConfig.load("configs/gpt2_pretrain.py") # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) get_data_from_cache(MERGE_FILE_URL, cache_dir, md5=MERGE_FILE_MD5) get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) get_data_from_cache(IDX_DATA_URL, cache_dir, md5=IDX_DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() vocab_path = get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) merges_file = get_data_from_cache(MERGE_FILE_URL, cache_dir, md5=MERGE_FILE_MD5) data_prefix_path = get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) data_prefix = data_prefix_path[:-4] # set tokenizer and data config cfg.tokenization.tokenizer.vocab_file = vocab_path cfg.tokenization.tokenizer.merges_file = merges_file cfg.dataloader.train.dataset[0].data_prefix = data_prefix cfg.dataloader.train.dataset[0].indexed_dataset.data_prefix = data_prefix cfg.dataloader.train.num_workers = 0 # set training config cfg.train.train_epoch = 0 cfg.train.train_iter = 10 cfg.train.evaluation.enabled = True cfg.train.evaluation.eval_period = 10 cfg.train.evaluation.eval_iter = 10 cfg.train.log_period = 1 cfg.train.train_micro_batch_size = 4 cfg.train.num_accumulation_steps = 1 cfg.train.resume = False cfg.train.output_dir = TEST_OUTPUT # set model cfg.model.cfg.max_seq_length = 256 cfg.model.cfg.num_attention_heads = 8 cfg.model.cfg.hidden_size = 384 cfg.model.cfg.hidden_layers = 4 cfg.model.cfg.hidden_layers = 4 cfg.train.activation_checkpoint.enabled = True cfg.train.amp.enabled = True cfg.train.rdma_enabled = False for ds in cfg.dataloader.train.dataset: ds.max_seq_length = cfg.model.cfg.max_seq_length self.cfg = cfg def build_hooks(self): ret = [ hooks.IterationTimer(), hooks.LRScheduler(), ] if dist.is_main_process(): # run writers in the end, so that evaluation metrics are written ret.append(hooks.PeriodicWriter(self.build_writers(), self.cfg.train.log_period)) return ret DefaultTrainer.build_hooks = build_hooks @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_gpt_eager_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 # pipeline parallelism not supported in eager global now! self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_gpt_eager_with_pipeline_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 1 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 4 self.cfg.train.dist.pipeline_num_layers = self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_gpt_graph_with_data_tensor_parallel(self): self.cfg.train.num_accumulation_steps = 1 # set distributed config self.cfg.train.dist.data_parallel_size = 4 self.cfg.train.dist.tensor_parallel_size = 2 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_gpt_graph_with_data_tensor_pipeline_parallel(self): self.cfg.train.num_accumulation_steps = 4 # set distributed config self.cfg.train.dist.data_parallel_size = 2 # change to 2 when 2d sbp bugfix self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 2 self.cfg.train.dist.pipeline_num_layers = self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_gpt_with_zero(self): # set distributed config self.cfg.train.dist.data_parallel_size = 4 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True self.cfg.train.zero_optimization.enabled = True self.cfg.train.zero_optimization.stage = 3 trainer = DefaultTrainer(self.cfg) trainer.train() if __name__ == "__main__": unittest.main()
7,701
36.754902
145
py
libai
libai-main/tests/models/test_t5.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import oneflow as flow import oneflow.unittest from libai.config import LazyConfig from libai.engine import DefaultTrainer, hooks from libai.engine.default import _check_batch_size from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger VOCAB_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/bert-base-chinese-vocab.txt" # noqa BIN_DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/loss_compara_content_sentence.bin" # noqa IDX_DATA_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/loss_compara_content_sentence.idx" # noqa VOCAB_MD5 = "3b5b76c4aef48ecf8cb3abaafe960f09" BIN_DATA_MD5 = "b842467bd5ea7e52f7a612ea6b4faecc" IDX_DATA_MD5 = "cf5963b8543f0a7a867361eb980f0372" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_t5") setup_logger(distributed_rank=dist.get_rank()) class TestT5Model(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join(os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "bert_data") cfg = LazyConfig.load("configs/t5_large_pretrain.py") # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) get_data_from_cache(IDX_DATA_URL, cache_dir, md5=IDX_DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() vocab_path = get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) data_prefix_path = get_data_from_cache(BIN_DATA_URL, cache_dir, md5=BIN_DATA_MD5) data_prefix = data_prefix_path[:-4] # set tokenizer and data config cfg.tokenization.tokenizer.vocab_file = vocab_path cfg.dataloader.train.dataset[0].data_prefix = data_prefix cfg.dataloader.train.dataset[0].indexed_dataset.data_prefix = data_prefix # FIXME(RenTianhe): fix dataloader worker bug cfg.dataloader.train.num_workers = 0 # set training config cfg.train.train_epoch = 0 cfg.train.train_iter = 10 cfg.train.evaluation.enabled = True cfg.train.evaluation.eval_period = 10 cfg.train.evaluation.eval_iter = 10 cfg.train.log_period = 1 cfg.train.train_micro_batch_size = 8 cfg.train.num_accumulation_steps = 1 cfg.train.resume = False cfg.train.output_dir = TEST_OUTPUT # set model cfg.model.cfg.num_attention_heads = 8 cfg.model.cfg.hidden_size = 384 cfg.model.cfg.hidden_layers = 3 cfg.train.activation_checkpoint.enabled = True cfg.train.amp.enabled = True cfg.train.rdma_enabled = False self.cfg = cfg def build_hooks(self): ret = [ hooks.IterationTimer(), hooks.LRScheduler(), ] if dist.is_main_process(): # run writers in the end, so that evaluation metrics are written ret.append(hooks.PeriodicWriter(self.build_writers(), self.cfg.train.log_period)) return ret DefaultTrainer.build_hooks = build_hooks @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_t5_eager_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 # pipeline parallelism not supported in eager global now! self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_t5_eager_with_pipeline_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 1 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 4 self.cfg.train.dist.pipeline_num_layers = 2 * self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = False trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_t5_graph_with_data_tensor_parallel(self): # set distributed config self.cfg.train.dist.data_parallel_size = 2 self.cfg.train.dist.tensor_parallel_size = 2 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_t5_graph_with_data_tensor_pipeline_parallel(self): self.cfg.train.num_accumulation_steps = 4 # set distributed config self.cfg.train.dist.data_parallel_size = 2 # change to 2 when 2d sbp bugfix self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 2 # encoder_layers + decoder_layers self.cfg.train.dist.pipeline_num_layers = 2 * self.cfg.model.cfg.hidden_layers dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True trainer = DefaultTrainer(self.cfg) trainer.train() @flow.unittest.skip_unless_1n4d() def test_t5_with_zero(self): # set distributed config self.cfg.train.dist.data_parallel_size = 4 self.cfg.train.dist.tensor_parallel_size = 1 self.cfg.train.dist.pipeline_parallel_size = 1 dist.setup_dist_util(self.cfg.train.dist) _check_batch_size(self.cfg) self.cfg.graph.enabled = True self.cfg.train.zero_optimization.enabled = True self.cfg.train.zero_optimization.stage = 3 trainer = DefaultTrainer(self.cfg) trainer.train() if __name__ == "__main__": unittest.main()
7,161
36.108808
145
py
libai
libai-main/tests/tokenizer/test_tokenization_roberta.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import json import os import unittest from libai.tokenizer.tokenization_roberta import VOCAB_FILES_NAMES, RobertaTokenizer from tests.tokenizer.test_tokenization_common import TokenizerTesterMixin class RobertaTokenizationTest(TokenizerTesterMixin, unittest.TestCase): tokenizer_class = RobertaTokenizer def setUp(self): super().setUp() # Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt vocab = [ "l", "o", "w", "e", "r", "s", "t", "i", "d", "n", "\u0120", "\u0120l", "\u0120n", "\u0120lo", "\u0120low", "er", "\u0120lowest", "\u0120newer", "\u0120wider", "<unk>", ] vocab_tokens = dict(zip(vocab, range(len(vocab)))) merges = ["#version: 0.2", "\u0120 l", "\u0120l o", "\u0120lo w", "e r", ""] self.special_tokens_map = {"unk_token": "<unk>"} self.vocab_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["vocab_file"]) self.merges_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["merges_file"]) with open(self.vocab_file, "w", encoding="utf-8") as fp: fp.write(json.dumps(vocab_tokens) + "\n") with open(self.merges_file, "w", encoding="utf-8") as fp: fp.write("\n".join(merges)) def get_tokenizer(self, **kwargs): kwargs.update(self.special_tokens_map) return RobertaTokenizer.from_pretrained(self.tmpdirname, **kwargs) def get_input_output_texts(self, tokenizer): input_text = "lower newer" output_text = "lower newer" return input_text, output_text def test_full_tokenizer(self): tokenizer = RobertaTokenizer(self.vocab_file, self.merges_file, **self.special_tokens_map) text = "lower newer" bpe_tokens = ["l", "o", "w", "er", "\u0120", "n", "e", "w", "er"] tokens = tokenizer.tokenize(text) self.assertListEqual(tokens, bpe_tokens) input_tokens = tokens + [tokenizer.unk_token] input_bpe_tokens = [0, 1, 2, 15, 10, 9, 3, 2, 15, 19] self.assertListEqual(tokenizer.convert_tokens_to_ids(input_tokens), input_bpe_tokens) def roberta_dict_integration_testing(self): tokenizer = self.get_tokenizer() self.assertListEqual( tokenizer.encode("Hello world!", add_special_tokens=False), [0, 31414, 232, 328, 2] ) self.assertListEqual( tokenizer.encode("Hello world! cécé herlolip 418", add_special_tokens=False), [0, 31414, 232, 328, 740, 1140, 12695, 69, 46078, 1588, 2], ) if __name__ == "__main__": unittest.main()
3,451
34.22449
98
py
libai
libai-main/tests/tokenizer/test_tokenization_common.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import inspect import os import pickle import re import shutil import tempfile from typing import Tuple from libai.tokenizer import PreTrainedTokenizer from tests.fixtures.utils import get_fixtures def get_tests_dir(append_path=None): """ Args: append_path: optional path to append to the tests dir path Return: The full path to the `tests` dir, so that the tests can be invoked from anywhere. Optionally `append_path` is joined after the `tests` dir the former is provided. """ # this function caller's __file__ caller__file__ = inspect.stack()[1][1] tests_dir = os.path.abspath(os.path.dirname(caller__file__)) if append_path: return os.path.join(tests_dir, append_path) else: return tests_dir class TokenizerTesterMixin: tokenizer_class = None def setUp(self): self.tokenizers_list = [] get_fixtures("sample_text.txt") with open(f"{get_tests_dir()}/../fixtures/sample_text.txt", encoding="utf-8") as f_data: self._data = f_data.read().replace("\n\n", "\n").strip() self.tmpdirname = tempfile.mkdtemp() def tearDown(self): shutil.rmtree(self.tmpdirname) def get_input_output_texts(self, tokenizer): input_txt = self.get_clean_sequence(tokenizer)[0] return input_txt, input_txt def get_clean_sequence( self, tokenizer, with_prefix_space=False, max_length=20 ) -> Tuple[str, list]: toks = [ (i, tokenizer.decode([i], clean_up_tokenization_spaces=False)) for i in range(len(tokenizer)) ] toks = list(filter(lambda t: re.match(r"^[ a-zA-Z]+$", t[1]), toks)) toks = list(filter(lambda t: [t[0]] == tokenizer.encode(t[1]), toks)) if max_length is not None and len(toks) > max_length: toks = toks[:max_length] toks_ids = [t[0] for t in toks] # Ensure consistency output_txt = tokenizer.decode(toks_ids, clean_up_tokenization_spaces=False) if " " not in output_txt and len(toks_ids) > 1: output_txt = ( tokenizer.decode([toks_ids[0]], clean_up_tokenization_spaces=False) + " " + tokenizer.decode(toks_ids[1:], clean_up_tokenization_spaces=False) ) if with_prefix_space: output_txt = " " + output_txt output_ids = tokenizer.encode(output_txt) return output_txt, output_ids def get_tokenizers(self, **kwargs): return [self.get_tokenizer(**kwargs)] def get_tokenizer(self, **kwargs) -> PreTrainedTokenizer: return self.tokenizer_class.from_pretrained(self.tmpdirname, **kwargs) def test_tokenizers_common_properties(self): tokenizers = self.get_tokenizers() for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): attributes_list = [ "bos_token", "eos_token", "unk_token", "sep_token", "pad_token", "cls_token", "mask_token", ] for attr in attributes_list: self.assertTrue(hasattr(tokenizer, attr)) self.assertTrue(hasattr(tokenizer, attr + "_id")) self.assertTrue(hasattr(tokenizer, "additional_special_tokens")) self.assertTrue(hasattr(tokenizer, "additional_special_tokens_ids")) attributes_list = [ "init_inputs", "init_kwargs", "added_tokens_encoder", "added_tokens_decoder", ] for attr in attributes_list: self.assertTrue(hasattr(tokenizer, attr)) def test_save_and_load_tokenizer(self): # Now let's start the test tokenizers = self.get_tokenizers() for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): # Isolate this from the other tests because we save additional tokens/etc tmpdirname = tempfile.mkdtemp() sample_text = " He is very happy, UNwant\u00E9d,running" before_tokens = tokenizer.encode(sample_text) before_vocab = tokenizer.get_vocab() tokenizer.save_pretrained(tmpdirname) after_tokenizer = tokenizer.__class__.from_pretrained(tmpdirname) after_tokens = after_tokenizer.encode(sample_text) after_vocab = after_tokenizer.get_vocab() self.assertListEqual(before_tokens, after_tokens) self.assertDictEqual(before_vocab, after_vocab) shutil.rmtree(tmpdirname) # Now let's start the test tokenizers = self.get_tokenizers() for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): # Isolate this from the other tests because we save additional tokens/etc tmpdirname = tempfile.mkdtemp() sample_text = " He is very happy, UNwant\u00E9d,running" tokenizer.add_tokens(["bim", "bambam"]) additional_special_tokens = tokenizer.additional_special_tokens additional_special_tokens.append("new_additional_special_token") tokenizer.add_special_tokens( {"additional_special_tokens": additional_special_tokens} ) before_tokens = tokenizer.encode(sample_text) before_vocab = tokenizer.get_vocab() tokenizer.save_pretrained(tmpdirname) after_tokenizer = tokenizer.__class__.from_pretrained(tmpdirname) after_tokens = after_tokenizer.encode(sample_text) after_vocab = after_tokenizer.get_vocab() self.assertListEqual(before_tokens, after_tokens) self.assertDictEqual(before_vocab, after_vocab) self.assertIn("bim", after_vocab) self.assertIn("bambam", after_vocab) self.assertIn( "new_additional_special_token", after_tokenizer.additional_special_tokens ) shutil.rmtree(tmpdirname) def test_pickle_tokenizer(self): """Google pickle __getstate__ __setstate__ if you are struggling with this.""" tokenizers = self.get_tokenizers() for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): self.assertIsNotNone(tokenizer) text = "Munich and Berlin are nice cities" subwords = tokenizer.tokenize(text) filename = os.path.join(self.tmpdirname, "tokenizer.bin") with open(filename, "wb") as handle: pickle.dump(tokenizer, handle) with open(filename, "rb") as handle: tokenizer_new = pickle.load(handle) subwords_loaded = tokenizer_new.tokenize(text) self.assertListEqual(subwords, subwords_loaded) def test_added_tokens_do_lower_case(self): tokenizers = self.get_tokenizers(do_lower_case=True) for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): if not hasattr(tokenizer, "do_lower_case") or not tokenizer.do_lower_case: continue special_token = tokenizer.all_special_tokens[0] text = special_token + " aaaaa bbbbbb low cccccccccdddddddd l " + special_token text2 = special_token + " AAAAA BBBBBB low CCCCCCCCCDDDDDDDD l " + special_token toks0 = tokenizer.tokenize(text) # toks before adding new_toks new_toks = [ "aaaaa bbbbbb", "cccccccccdddddddd", "AAAAA BBBBBB", "CCCCCCCCCDDDDDDDD", ] added = tokenizer.add_tokens(new_toks) self.assertEqual(added, 2) toks = tokenizer.tokenize(text) toks2 = tokenizer.tokenize(text2) self.assertEqual(len(toks), len(toks2)) self.assertListEqual(toks, toks2) self.assertNotEqual(len(toks), len(toks0)) # toks0 should be longer # Check that none of the special tokens are lowercased sequence_with_special_tokens = ( "A " + " yEs ".join(tokenizer.all_special_tokens) + " B" ) tokenized_sequence = tokenizer.tokenize(sequence_with_special_tokens) for special_token in tokenizer.all_special_tokens: self.assertTrue(special_token in tokenized_sequence) tokenizers = self.get_tokenizers(do_lower_case=False) for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): special_token = tokenizer.all_special_tokens[0] text = special_token + " aaaaa bbbbbb low cccccccccdddddddd l " + special_token text2 = special_token + " AAAAA BBBBBB low CCCCCCCCCDDDDDDDD l " + special_token new_toks = [ "aaaaa bbbbbb", "cccccccccdddddddd", "AAAAA BBBBBB", "CCCCCCCCCDDDDDDDD", ] toks0 = tokenizer.tokenize(text) # toks before adding new_toks added = tokenizer.add_tokens(new_toks) self.assertEqual(added, 4) toks = tokenizer.tokenize(text) toks2 = tokenizer.tokenize(text2) self.assertEqual(len(toks), len(toks2)) # Length should still be the same self.assertNotEqual( toks[1], toks2[1] ) # But at least the first non-special tokens should differ self.assertNotEqual(len(toks), len(toks0)) # toks0 should be longer def test_add_tokens_tokenizer(self): tokenizers = self.get_tokenizers(do_lower_case=False) for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): vocab_size = tokenizer.vocab_size all_size = len(tokenizer) self.assertNotEqual(vocab_size, 0) # We usually have added tokens from the start in tests # because our vocab fixtures are smaller than the original vocabs # let's not assert this self.assertEqual(vocab_size, all_size) new_toks = ["aaaaa bbbbbb", "cccccccccdddddddd"] added_toks = tokenizer.add_tokens(new_toks) vocab_size_2 = tokenizer.vocab_size all_size_2 = len(tokenizer) self.assertNotEqual(vocab_size_2, 0) self.assertEqual(vocab_size, vocab_size_2) self.assertEqual(added_toks, len(new_toks)) self.assertEqual(all_size_2, all_size + len(new_toks)) tokens = tokenizer.encode("aaaaa bbbbbb low cccccccccdddddddd l") self.assertGreaterEqual(len(tokens), 4) self.assertGreater(tokens[0], tokenizer.vocab_size - 1) self.assertGreater(tokens[-2], tokenizer.vocab_size - 1) new_toks_2 = {"eos_token": ">>>>|||<||<<|<<", "pad_token": "<<<<<|||>|>>>>|>"} added_toks_2 = tokenizer.add_special_tokens(new_toks_2) vocab_size_3 = tokenizer.vocab_size all_size_3 = len(tokenizer) self.assertNotEqual(vocab_size_3, 0) self.assertEqual(vocab_size, vocab_size_3) self.assertEqual(added_toks_2, len(new_toks_2)) self.assertEqual(all_size_3, all_size_2 + len(new_toks_2)) tokens = tokenizer.encode( ">>>>|||<||<<|<< aaaaabbbbbb low cccccccccdddddddd <<<<<|||>|>>>>|> l" ) self.assertGreaterEqual(len(tokens), 6) self.assertGreater(tokens[0], tokenizer.vocab_size - 1) self.assertGreater(tokens[0], tokens[1]) self.assertGreater(tokens[-2], tokenizer.vocab_size - 1) self.assertGreater(tokens[-2], tokens[-3]) self.assertEqual(tokens[0], tokenizer.eos_token_id) self.assertEqual(tokens[-2], tokenizer.pad_token_id) def test_add_special_tokens(self): tokenizers = self.get_tokenizers(do_lower_case=False) for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): input_text, ids = self.get_clean_sequence(tokenizer) special_token = "[SPECIAL_TOKEN]" tokenizer.add_special_tokens({"cls_token": special_token}) encoded_special_token = tokenizer.encode(special_token) self.assertEqual(len(encoded_special_token), 1) text = tokenizer.decode( ids + encoded_special_token, clean_up_tokenization_spaces=False ) encoded = tokenizer.encode(text) input_encoded = tokenizer.encode(input_text) special_token_id = tokenizer.encode(special_token) self.assertEqual(encoded, input_encoded + special_token_id) decoded = tokenizer.decode(encoded, skip_special_tokens=True) self.assertTrue(special_token not in decoded) def test_internal_consistency(self): tokenizers = self.get_tokenizers() for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): input_text, output_text = self.get_input_output_texts(tokenizer) tokens = tokenizer.tokenize(input_text) ids = tokenizer.convert_tokens_to_ids(tokens) ids_2 = tokenizer.encode(input_text) self.assertListEqual(ids, ids_2) tokens_2 = tokenizer.convert_ids_to_tokens(ids) self.assertNotEqual(len(tokens_2), 0) text_2 = tokenizer.decode(ids) self.assertIsInstance(text_2, str) self.assertEqual(text_2, output_text) def test_encode_decode_with_spaces(self): tokenizers = self.get_tokenizers(do_lower_case=False) for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): new_toks = ["[ABC]", "[DEF]"] tokenizer.add_tokens(new_toks) input = "[ABC] [DEF] [ABC] [DEF]" encoded = tokenizer.encode(input) decoded = tokenizer.decode(encoded) self.assertEqual(decoded, input) def test_pretrained_model_lists(self): weights_list = list(self.tokenizer_class.max_model_input_sizes.keys()) weights_lists_2 = [] for file_id, map_list in self.tokenizer_class.pretrained_vocab_files_map.items(): weights_lists_2.append(list(map_list.keys())) for weights_list_2 in weights_lists_2: self.assertListEqual(weights_list, weights_list_2) def test_get_vocab(self): tokenizers = self.get_tokenizers(do_lower_case=False) for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): vocab = tokenizer.get_vocab() self.assertIsInstance(vocab, dict) self.assertEqual(len(vocab), len(tokenizer)) for word, ind in vocab.items(): self.assertEqual(tokenizer.convert_tokens_to_ids(word), ind) self.assertEqual(tokenizer.convert_ids_to_tokens(ind), word) tokenizer.add_tokens(["asdfasdfasdfasdf"]) vocab = tokenizer.get_vocab() self.assertIsInstance(vocab, dict) self.assertEqual(len(vocab), len(tokenizer))
16,794
40.882793
96
py
libai
libai-main/tests/tokenizer/test_tokenization_bert.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import unittest from libai.tokenizer.tokenization_base import _is_control, _is_punctuation, _is_whitespace from libai.tokenizer.tokenization_bert import ( VOCAB_FILES_NAMES, BasicTokenizer, BertTokenizer, WordpieceTokenizer, ) from tests.tokenizer.test_tokenization_common import TokenizerTesterMixin class BertTokenizationTest(TokenizerTesterMixin, unittest.TestCase): tokenizer_class = BertTokenizer def setUp(self): super().setUp() vocab_tokens = [ "[UNK]", "[CLS]", "[SEP]", "[PAD]", "[MASK]", "want", "##want", "##ed", "wa", "un", "runn", "##ing", ",", "low", "lowest", ] self.vocab_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["vocab_file"]) with open(self.vocab_file, "w", encoding="utf-8") as vocab_writer: vocab_writer.write("".join([x + "\n" for x in vocab_tokens])) def get_input_output_texts(self, tokenizer): input_text = "UNwant\u00E9d,running" output_text = "unwanted, running" return input_text, output_text def test_full_tokenizer(self): tokenizer = self.tokenizer_class(self.vocab_file) tokens = tokenizer.tokenize("UNwant\u00E9d,running") self.assertListEqual(tokens, ["un", "##want", "##ed", ",", "runn", "##ing"]) self.assertListEqual(tokenizer.convert_tokens_to_ids(tokens), [9, 6, 7, 12, 10, 11]) def test_chinese(self): tokenizer = BasicTokenizer() self.assertListEqual( tokenizer.tokenize("ah\u535A\u63A8zz"), ["ah", "\u535A", "\u63A8", "zz"] ) def test_basic_tokenizer_lower(self): tokenizer = BasicTokenizer(do_lower_case=True) self.assertListEqual( tokenizer.tokenize(" \tHeLLo!how \n Are yoU? "), ["hello", "!", "how", "are", "you", "?"], ) self.assertListEqual(tokenizer.tokenize("H\u00E9llo"), ["hello"]) def test_basic_tokenizer_no_lower(self): tokenizer = BasicTokenizer(do_lower_case=False) self.assertListEqual( tokenizer.tokenize(" \tHeLLo!how \n Are yoU? "), ["HeLLo", "!", "how", "Are", "yoU", "?"], ) def test_basic_tokenizer_respects_never_split_tokens(self): tokenizer = BasicTokenizer(do_lower_case=False, never_split=["[UNK]"]) self.assertListEqual( tokenizer.tokenize(" \tHeLLo!how \n Are yoU? [UNK]"), ["HeLLo", "!", "how", "Are", "yoU", "?", "[UNK]"], ) def test_wordpiece_tokenizer(self): vocab_tokens = [ "[UNK]", "[CLS]", "[SEP]", "want", "##want", "##ed", "wa", "un", "runn", "##ing", ] vocab = {} for (i, token) in enumerate(vocab_tokens): vocab[token] = i tokenizer = WordpieceTokenizer(vocab=vocab, unk_token="[UNK]") self.assertListEqual(tokenizer.tokenize(""), []) self.assertListEqual( tokenizer.tokenize("unwanted running"), ["un", "##want", "##ed", "runn", "##ing"] ) self.assertListEqual(tokenizer.tokenize("unwantedX running"), ["[UNK]", "runn", "##ing"]) def test_is_whitespace(self): self.assertTrue(_is_whitespace(" ")) self.assertTrue(_is_whitespace("\t")) self.assertTrue(_is_whitespace("\r")) self.assertTrue(_is_whitespace("\n")) self.assertTrue(_is_whitespace("\u00A0")) self.assertFalse(_is_whitespace("A")) self.assertFalse(_is_whitespace("-")) def test_is_control(self): self.assertTrue(_is_control("\u0005")) self.assertFalse(_is_control("A")) self.assertFalse(_is_control(" ")) self.assertFalse(_is_control("\t")) self.assertFalse(_is_control("\r")) def test_is_punctuation(self): self.assertTrue(_is_punctuation("-")) self.assertTrue(_is_punctuation("$")) self.assertTrue(_is_punctuation("`")) self.assertTrue(_is_punctuation(".")) self.assertFalse(_is_punctuation("A")) self.assertFalse(_is_punctuation(" ")) if __name__ == "__main__": unittest.main()
5,007
30.696203
97
py
libai
libai-main/tests/tokenizer/test_tokenization_gpt2.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import json import os import unittest from libai.tokenizer.tokenization_gpt2 import VOCAB_FILES_NAMES, GPT2Tokenizer from tests.tokenizer.test_tokenization_common import TokenizerTesterMixin class GPT2TokenizationTest(TokenizerTesterMixin, unittest.TestCase): tokenizer_class = GPT2Tokenizer def setUp(self): super().setUp() # Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt vocab = [ "l", "o", "w", "e", "r", "s", "t", "i", "d", "n", "\u0120", "\u0120l", "\u0120n", "\u0120lo", "\u0120low", "er", "\u0120lowest", "\u0120newer", "\u0120wider", "<unk>", "<|endoftext|>", ] vocab_tokens = dict(zip(vocab, range(len(vocab)))) merges = ["#version: 0.2", "\u0120 l", "\u0120l o", "\u0120lo w", "e r", ""] self.special_tokens_map = {"unk_token": "<unk>"} self.vocab_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["vocab_file"]) self.merges_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["merges_file"]) with open(self.vocab_file, "w", encoding="utf-8") as fp: fp.write(json.dumps(vocab_tokens) + "\n") with open(self.merges_file, "w", encoding="utf-8") as fp: fp.write("\n".join(merges)) def get_tokenizer(self, **kwargs): kwargs.update(self.special_tokens_map) return GPT2Tokenizer.from_pretrained(self.tmpdirname, **kwargs) def get_input_output_texts(self, tokenizer): input_text = "lower newer" output_text = "lower newer" return input_text, output_text def test_full_tokenizer(self): tokenizer = GPT2Tokenizer(self.vocab_file, self.merges_file, **self.special_tokens_map) text = " lower newer" bpe_tokens = ["\u0120low", "er", "\u0120", "n", "e", "w", "er"] tokens = tokenizer.tokenize(text) self.assertListEqual(tokens, bpe_tokens) input_tokens = tokens + [tokenizer.unk_token] input_bpe_tokens = [14, 15, 10, 9, 3, 2, 15, 19] self.assertListEqual(tokenizer.convert_tokens_to_ids(input_tokens), input_bpe_tokens) if __name__ == "__main__": unittest.main()
3,028
33.033708
95
py
libai
libai-main/tests/tokenizer/__init__.py
0
0
0
py
libai
libai-main/tests/tokenizer/test_tokenization_t5.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import tempfile import unittest from libai.tokenizer.tokenization_t5 import T5Tokenizer from tests.fixtures.utils import get_fixtures from tests.tokenizer.test_tokenization_common import TokenizerTesterMixin SPIECE_UNDERLINE = "▁" SAMPLE_VOCAB = os.path.join( os.path.dirname(os.path.abspath(__file__)), "../fixtures/test_sentencepiece.model" ) class T5TokenizationTest(TokenizerTesterMixin, unittest.TestCase): tokenizer_class = T5Tokenizer def setUp(self): super().setUp() # We have a SentencePiece fixture for testing get_fixtures(SAMPLE_VOCAB) tokenizer = T5Tokenizer(SAMPLE_VOCAB) tokenizer.save_pretrained(self.tmpdirname) def test_convert_token_and_id(self): """Test ``_convert_token_to_id`` and ``_convert_id_to_token``.""" token = "<s>" token_id = 1 self.assertEqual(self.get_tokenizer()._convert_token_to_id(token), token_id) self.assertEqual(self.get_tokenizer()._convert_id_to_token(token_id), token) def test_get_vocab(self): vocab_keys = list(self.get_tokenizer().get_vocab().keys()) self.assertEqual(vocab_keys[0], "<unk>") self.assertEqual(vocab_keys[1], "<s>") self.assertEqual(vocab_keys[-1], "<pad>") self.assertEqual(len(vocab_keys), 1_101) def test_vocab_size(self): self.assertEqual(self.get_tokenizer().vocab_size, 1_100) def test_full_tokenizer(self): tokenizer = T5Tokenizer(SAMPLE_VOCAB) tokens = tokenizer.tokenize("This is a test") self.assertListEqual(tokens, ["▁This", "▁is", "▁a", "▁t", "est"]) self.assertListEqual(tokenizer.convert_tokens_to_ids(tokens), [285, 46, 10, 170, 382]) tokens = tokenizer.tokenize("I was born in 92000, and this is falsé.") self.assertListEqual( tokens, [ SPIECE_UNDERLINE + "I", SPIECE_UNDERLINE + "was", SPIECE_UNDERLINE + "b", "or", "n", SPIECE_UNDERLINE + "in", SPIECE_UNDERLINE + "", "9", "2", "0", "0", "0", ",", SPIECE_UNDERLINE + "and", SPIECE_UNDERLINE + "this", SPIECE_UNDERLINE + "is", SPIECE_UNDERLINE + "f", "al", "s", "é", ".", ], ) ids = tokenizer.convert_tokens_to_ids(tokens) self.assertListEqual( ids, [8, 21, 84, 55, 24, 19, 7, 0, 602, 347, 347, 347, 3, 12, 66, 46, 72, 80, 6, 0, 4] ) back_tokens = tokenizer.convert_ids_to_tokens(ids) self.assertListEqual( back_tokens, [ SPIECE_UNDERLINE + "I", SPIECE_UNDERLINE + "was", SPIECE_UNDERLINE + "b", "or", "n", SPIECE_UNDERLINE + "in", SPIECE_UNDERLINE + "", "<unk>", "2", "0", "0", "0", ",", SPIECE_UNDERLINE + "and", SPIECE_UNDERLINE + "this", SPIECE_UNDERLINE + "is", SPIECE_UNDERLINE + "f", "al", "s", "<unk>", ".", ], ) def test_save_and_load_tokenizer(self): # Now let's start the test tokenizers = self.get_tokenizers() for tokenizer in tokenizers: with self.subTest(f"{tokenizer.__class__.__name__}"): # Isolate this from the other tests because we save additional tokens/etc tmpdirname = tempfile.mkdtemp() sample_text = " He is very happy, UNwant\u00E9d,running" before_tokens = tokenizer.encode(sample_text) before_vocab = tokenizer.get_vocab() tokenizer.save_pretrained(tmpdirname) after_tokenizer = tokenizer.__class__.from_pretrained(tmpdirname) after_tokens = after_tokenizer.encode(sample_text) after_vocab = after_tokenizer.get_vocab() self.assertListEqual(before_tokens, after_tokens) self.assertDictEqual(before_vocab, after_vocab) shutil.rmtree(tmpdirname) if __name__ == "__main__": unittest.main()
5,168
32.134615
98
py
libai
libai-main/tests/model_loader/test_mt5_loader.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import numpy as np import oneflow as flow import oneflow.unittest from omegaconf import DictConfig from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger from projects.MT5.configs.mt5_base import cfg as libai_cfg from projects.MT5.mt5_model import MT5Model from projects.MT5.utils.mt5_loader import T5LoaderHuggerFace PRETRAINED_MODEL_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/mt5_utils/pytorch_model.bin" # noqa PRETRAINED_MODEL_CONFIG_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/mt5_utils/config.json" # noqa PRETRAINED_MODEL_MD5 = "4c9c0be541b89de9b01c597ec4cc371a" PRETRAINED_MODEL_CONFIG_MD5 = "b159e41603b7eeaf9a9c489165bbcaca" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_mt5_utils") setup_logger(distributed_rank=dist.get_rank()) class TestMT5Loader(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join( os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "mt5_utils_data" ) self.pretrained_model_path = cache_dir # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(PRETRAINED_MODEL_URL, cache_dir, md5=PRETRAINED_MODEL_MD5) get_data_from_cache( PRETRAINED_MODEL_CONFIG_URL, cache_dir, md5=PRETRAINED_MODEL_CONFIG_MD5 ) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() # prepare input data self.encoder_input_ids = [ [101, 2009, 1005, 1055, 2986, 2651, 1012, 102], [101, 2028, 12314, 3377, 102, 0, 0, 0], [101, 2064, 2017, 3305, 2009, 102, 0, 0], ] self.encoder_att_mask = [ [1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1], ] self.decoder_input_ids = [ [101, 2009, 1005, 1055, 2986], [101, 2028, 12314, 3377, 102], [101, 2064, 2017, 3305, 2009], ] self.decoder_att_mask = [[1, 1, 1, 1, 1], [1, 1, 1, 1, 1], [1, 1, 1, 1, 1]] @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_mt5_loader_with_data_tensor_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = T5LoaderHuggerFace( model=MT5Model, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, hidden_dropout_prob=0.0, attention_probs_dropout_prob=0.0, embedding_dropout_prob=0.0, model_type="mt5", ) model = load_func.load() model.eval() encoder_input_ids = flow.tensor( self.encoder_input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) decoder_input_ids = flow.tensor( self.decoder_input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) encode_att_mask = flow.tensor( self.encoder_att_mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) decoder_att_mask = flow.tensor( self.decoder_att_mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) logits = model( encoder_input_ids, decoder_input_ids, encode_att_mask, decoder_att_mask, encode_att_mask )["logits"] self.assertTrue( np.allclose( np.array(-83584720.0), logits.sum().data.numpy(), ) ) @flow.unittest.skip_unless_1n4d() def test_mt5_loader_with_data_tensor_pipeline_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=16, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = T5LoaderHuggerFace( model=MT5Model, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, hidden_dropout_prob=0.0, attention_probs_dropout_prob=0.0, embedding_dropout_prob=0.0, model_type="mt5", ) model = load_func.load() model.eval() encoder_input_ids = flow.tensor( self.encoder_input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) decoder_input_ids = flow.tensor( self.decoder_input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) encode_att_mask = flow.tensor( self.encoder_att_mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) decoder_att_mask = flow.tensor( self.decoder_att_mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) logits = model( encoder_input_ids, decoder_input_ids, encode_att_mask, decoder_att_mask, encode_att_mask )["logits"] self.assertTrue( np.allclose( np.array(-83584720.0), logits.sum().data.numpy(), ) ) if __name__ == "__main__": unittest.main()
7,293
33.899522
151
py
libai
libai-main/tests/model_loader/test_t5_loader.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import numpy as np import oneflow as flow import oneflow.unittest from omegaconf import DictConfig from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger from projects.MT5.configs.mt5_base import cfg as libai_cfg from projects.MT5.mt5_model import MT5Model from projects.MT5.utils.mt5_loader import T5LoaderHuggerFace PRETRAINED_MODEL_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/t5_utils/pytorch_model.bin" # noqa PRETRAINED_MODEL_CONFIG_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/t5_utils/config.json" # noqa PRETRAINED_MODEL_MD5 = "952862a8ba425a25739a69e5f33b0df8" PRETRAINED_MODEL_CONFIG_MD5 = "7ebc91dc4377c01190f4116c3c1ac6cd" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_t5_utils") setup_logger(distributed_rank=dist.get_rank()) class TestT5Loader(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join( os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "t5_utils_data" ) self.pretrained_model_path = cache_dir # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(PRETRAINED_MODEL_URL, cache_dir, md5=PRETRAINED_MODEL_MD5) get_data_from_cache( PRETRAINED_MODEL_CONFIG_URL, cache_dir, md5=PRETRAINED_MODEL_CONFIG_MD5 ) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() # prepare input data self.encoder_input_ids = [ [101, 2009, 1005, 1055, 2986, 2651, 1012, 102], [101, 2028, 12314, 3377, 102, 0, 0, 0], [101, 2064, 2017, 3305, 2009, 102, 0, 0], ] self.encoder_att_mask = [ [1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1], ] self.decoder_input_ids = [ [101, 2009, 1005, 1055, 2986], [101, 2028, 12314, 3377, 102], [101, 2064, 2017, 3305, 2009], ] self.decoder_att_mask = [[1, 1, 1, 1, 1], [1, 1, 1, 1, 1], [1, 1, 1, 1, 1]] @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_t5_loader_with_data_tensor_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = T5LoaderHuggerFace( model=MT5Model, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, hidden_dropout_prob=0.0, attention_probs_dropout_prob=0.0, embedding_dropout_prob=0.0, model_type="t5", ) model = load_func.load() model.eval() encoder_input_ids = flow.tensor( self.encoder_input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) decoder_input_ids = flow.tensor( self.decoder_input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) encode_att_mask = flow.tensor( self.encoder_att_mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) decoder_att_mask = flow.tensor( self.decoder_att_mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) logits = model( encoder_input_ids, decoder_input_ids, encode_att_mask, decoder_att_mask, encode_att_mask )["logits"] self.assertTrue( np.allclose( np.array(-9836561.0), logits.sum().data.numpy(), ) ) @flow.unittest.skip_unless_1n4d() def test_t5_loader_with_data_tensor_pipeline_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=24, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = T5LoaderHuggerFace( model=MT5Model, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, hidden_dropout_prob=0.0, attention_probs_dropout_prob=0.0, embedding_dropout_prob=0.0, model_type="t5", ) model = load_func.load() model.eval() encoder_input_ids = flow.tensor( self.encoder_input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) decoder_input_ids = flow.tensor( self.decoder_input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) encode_att_mask = flow.tensor( self.encoder_att_mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) decoder_att_mask = flow.tensor( self.decoder_att_mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=dist.get_layer_placement(0), ) logits = model( encoder_input_ids, decoder_input_ids, encode_att_mask, decoder_att_mask, encode_att_mask )["logits"] self.assertTrue( np.allclose( np.array(-9836561.0), logits.sum().data.numpy(), ) ) if __name__ == "__main__": unittest.main()
7,282
33.84689
150
py
libai
libai-main/tests/model_loader/test_roberta_loader.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import numpy as np import oneflow as flow import oneflow.unittest from omegaconf import DictConfig import libai from configs.common.models.roberta import cfg as libai_cfg from libai.models.utils import RobertaLoaderHuggerFace from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger PRETRAINED_MODEL_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/roberta_utils/pytorch_model.bin" # noqa PRETRAINED_MODEL_CONFIG_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/roberta_utils/config.json" # noqa PRETRAINED_MODEL_MD5 = "73db58b6c51b028e0ee031f12261b51d" # noqa PRETRAINED_MODEL_CONFIG_MD5 = "a53c22291c7f25d5077260ad5ca4d5fa" # noqa TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_roberta_utils") setup_logger(distributed_rank=dist.get_rank()) class TestRobertaLoader(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join( os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "roberta_utils_data" ) self.pretrained_model_path = cache_dir # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(PRETRAINED_MODEL_URL, cache_dir, md5=PRETRAINED_MODEL_MD5) get_data_from_cache( PRETRAINED_MODEL_CONFIG_URL, cache_dir, md5=PRETRAINED_MODEL_CONFIG_MD5 ) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() # prepare input data self.input_ids = [ [101, 2009, 1005, 1055, 2986, 2651, 1012, 102], [101, 2028, 12314, 3377, 102, 0, 0, 0], [101, 2064, 2017, 3305, 2009, 102, 0, 0], ] self.mask = [[1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1]] @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_roberta_loader_with_data_tensor_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = RobertaLoaderHuggerFace( model=libai.models.RobertaModel, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=False, apply_query_key_layer_scaling=False, apply_residual_post_layernorm=True, amp_enabled=False, ) model = load_func.load() model.eval() input_ids = flow.tensor( self.input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.vocab_embeddings.weight.placement, ) mask = flow.tensor( self.mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.vocab_embeddings.weight.placement, ) last_hidden_state, _ = model(input_ids, mask) self.assertTrue( np.allclose(np.array(341.5831), last_hidden_state.sum().data.numpy(), 1e-4, 1e-4) ) @flow.unittest.skip_unless_1n4d() def test_roberta_loader_with_data_tensor_pipeline_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = RobertaLoaderHuggerFace( model=libai.models.RobertaModel, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=False, apply_query_key_layer_scaling=False, apply_residual_post_layernorm=True, amp_enabled=False, ) model = load_func.load() model.eval() input_ids = flow.tensor( self.input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.vocab_embeddings.weight.placement, ) mask = flow.tensor( self.mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.vocab_embeddings.weight.placement, ) last_hidden_state, _ = model(input_ids, mask) self.assertTrue( np.allclose(np.array(341.5831), last_hidden_state.sum().data.numpy(), 1e-4, 1e-4) ) if __name__ == "__main__": unittest.main()
6,023
34.64497
155
py
libai
libai-main/tests/model_loader/test_gpt_loader.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import numpy as np import oneflow as flow import oneflow.unittest from omegaconf import DictConfig import libai from configs.common.models.gpt import cfg as libai_cfg from libai.models.utils import GPT2LoaderHuggerFace from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger PRETRAINED_MODEL_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/gpt_utils/pytorch_model.bin" # noqa PRETRAINED_MODEL_CONFIG_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/gpt_utils/config.json" # noqa PRETRAINED_MODEL_MD5 = "c086214036308afc71896da17ca0442a" PRETRAINED_MODEL_CONFIG_MD5 = "6e1dba197b511b8759d6ad4551095a29" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_gpt_utils") setup_logger(distributed_rank=dist.get_rank()) class TestGPT2Loader(flow.unittest.TestCase): """The activation function of gpt2 in LiBai is GELU, so the result here is to replace the activation function of gpt2 in huggingface from gelu_new to gelu. """ def setUp(self) -> None: cache_dir = os.path.join( os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "gpt_utils_data" ) self.pretrained_model_path = cache_dir # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(PRETRAINED_MODEL_URL, cache_dir, md5=PRETRAINED_MODEL_MD5) get_data_from_cache( PRETRAINED_MODEL_CONFIG_URL, cache_dir, md5=PRETRAINED_MODEL_CONFIG_MD5 ) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() # prepare input data self.input_ids = [ [101, 2009, 1005, 1055, 2986, 2651, 1012, 102], [101, 2028, 12314, 3377, 102, 0, 0, 0], [101, 2064, 2017, 3305, 2009, 102, 0, 0], ] @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_gpt_loader_with_data_tensor_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = GPT2LoaderHuggerFace( model=libai.models.GPTModel, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=True, apply_query_key_layer_scaling=True, apply_residual_post_layernorm=False, amp_enabled=False, attention_dropout_prob=0, output_dropout_prob=0, ) model = load_func.load() model.eval() input_ids = flow.tensor( self.input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.token_embeddings.weight.placement, ) logits = model(input_ids) self.assertTrue( np.allclose( np.array(-93505050.0), logits.sum().data.numpy(), ) ) @flow.unittest.skip_unless_1n4d() def test_gpt_loader_with_data_tensor_pipeline_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = GPT2LoaderHuggerFace( model=libai.models.GPTModel, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=True, apply_query_key_layer_scaling=True, apply_residual_post_layernorm=False, amp_enabled=False, attention_dropout_prob=0, output_dropout_prob=0, ) model = load_func.load() model.eval() input_ids = flow.tensor( self.input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.token_embeddings.weight.placement, ) logits = model(input_ids) self.assertTrue( np.allclose( np.array(-93505050.0), logits.sum().data.numpy(), ) ) @flow.unittest.skip_unless_1n4d() def test_gpt_loader_with_data_tensor_parallel_backward(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = GPT2LoaderHuggerFace( model=libai.models.GPTModel, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=True, apply_query_key_layer_scaling=True, apply_residual_post_layernorm=False, amp_enabled=False, attention_dropout_prob=0, output_dropout_prob=0, embedding_dropout_prob=0, ) model = load_func.load() input_ids = flow.tensor( self.input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.token_embeddings.weight.placement, ) logits = model(input_ids) loss = logits.sum() loss.backward() self.assertTrue( np.allclose(-24882176.0, model.transformer.layernorm_f.weight.grad.sum().numpy()) ) self.assertTrue( np.allclose( 3.1779e08, model.embeddings.token_embeddings.weight.grad.sum().numpy(), 1e-3 ) ) @flow.unittest.skip_unless_1n4d() def test_gpt_loader_with_data_tensor_pipeline_parallel_backward(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = GPT2LoaderHuggerFace( model=libai.models.GPTModel, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=True, apply_query_key_layer_scaling=True, apply_residual_post_layernorm=False, amp_enabled=False, attention_dropout_prob=0, output_dropout_prob=0, embedding_dropout_prob=0, ) model = load_func.load() input_ids = flow.tensor( self.input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.token_embeddings.weight.placement, ) logits = model(input_ids) loss = logits.sum() loss.backward() self.assertTrue( np.allclose(-24882176.0, model.transformer.layernorm_f.weight.grad.sum().numpy()) ) self.assertTrue( np.allclose(317785760.0, model.embeddings.token_embeddings.weight.grad.sum().numpy()) ) if __name__ == "__main__": unittest.main()
8,936
32.724528
151
py
libai
libai-main/tests/model_loader/test_swin_loader.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import numpy as np import oneflow as flow import oneflow.unittest from omegaconf import DictConfig import libai from configs.common.models.swin.swin_tiny_patch4_window7_224 import cfg as libai_cfg from libai.models.utils import SwinLoaderHuggerFace from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger PRETRAINED_MODEL_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/swin_utils/pytorch_model.bin" # noqa PRETRAINED_MODEL_CONFIG_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/swin_utils/config.json" # noqa INIT_DATA = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/swin_utils/init_data.npz" # noqa PRETRAINED_MODEL_MD5 = "cd8c03d9cd4a9c536a5a245f663035b6" PRETRAINED_MODEL_CONFIG_MD5 = "a8a71ed22b99323edd6a1457bede5819" INIT_DATA_MD5 = "5fecdcd8d46bfefa310d19e084bd4815" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_swin_utils") setup_logger(distributed_rank=dist.get_rank()) class TestSwinLoder(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join( os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "swin_utils_data" ) self.pretrained_model_path = cache_dir self.init_data_path = os.path.join(cache_dir, "init_data.npz") # download model and data if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(PRETRAINED_MODEL_URL, cache_dir, md5=PRETRAINED_MODEL_MD5) get_data_from_cache( PRETRAINED_MODEL_CONFIG_URL, cache_dir, md5=PRETRAINED_MODEL_CONFIG_MD5 ) get_data_from_cache(INIT_DATA, cache_dir, md5=INIT_DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() # prepare input data self.input_image = np.load(self.init_data_path)["arr_0"] @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_swin_loader_with_data_tensor_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = SwinLoaderHuggerFace( model=libai.models.SwinTransformer, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, ) model = load_func.load() model.eval() input_image = flow.tensor( self.input_image.tolist(), dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] self.assertTrue(np.allclose(np.array(80.9373), prediction_scores.sum().data.numpy(), 1e-3)) @flow.unittest.skip_unless_1n4d() def test_swin_loader_with_data_tensor_pipeline_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = SwinLoaderHuggerFace( model=libai.models.SwinTransformer, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, ) model = load_func.load() model.eval() input_image = flow.tensor( self.input_image, dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] self.assertTrue(np.allclose(np.array(80.9373), prediction_scores.sum().data.numpy(), 1e-3)) @flow.unittest.skip_unless_1n4d() def test_swin_loader_with_data_tensor_parallel_backward(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = SwinLoaderHuggerFace( model=libai.models.SwinTransformer, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, drop_rate=0.0, drop_path_rate=0.0, ) model = load_func.load() input_image = flow.tensor( self.input_image, dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] loss = prediction_scores.sum() loss.backward() self.assertTrue(np.allclose(108775.88, model.head.weight.grad.sum().numpy(), 1e-3)) self.assertTrue( np.allclose(24.320518, model.patch_embed.norm.weight.grad.sum().numpy(), 1e-2) ) @flow.unittest.skip_unless_1n4d() def test_swin_loader_with_data_tensor_pipeline_parallel_backward(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = SwinLoaderHuggerFace( model=libai.models.SwinTransformer, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, drop_rate=0.0, drop_path_rate=0.0, ) model = load_func.load() input_image = flow.tensor( self.input_image, dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] loss = prediction_scores.sum() loss.backward() self.assertTrue(np.allclose(108775.88, model.head.weight.grad.sum().numpy(), 1e-3)) self.assertTrue( np.allclose(24.320518, model.patch_embed.norm.weight.grad.sum().numpy(), 1e-2) ) if __name__ == "__main__": unittest.main()
7,713
34.223744
152
py
libai
libai-main/tests/model_loader/test_vit_loader.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import numpy as np import oneflow as flow import oneflow.unittest from omegaconf import DictConfig import libai from configs.common.models.vit.vit_tiny_patch16_224 import cfg as libai_cfg from libai.models.utils import ViTLoaderHuggerFace from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger PRETRAINED_MODEL_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/vit_utils/pytorch_model.bin" # noqa PRETRAINED_MODEL_CONFIG_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/vit_utils/config.json" # noqa INIT_DATA = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/vit_utils/init_data.npz" # noqa PRETRAINED_MODEL_MD5 = "c587693e5e312064c56f27aa2d4f1e81" PRETRAINED_MODEL_CONFIG_MD5 = "9ea94d9e5bc3543b1de7d12956321c50" INIT_DATA_MD5 = "5fecdcd8d46bfefa310d19e084bd4815" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_vit_utils") setup_logger(distributed_rank=dist.get_rank()) class TestViTLoder(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join( os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "vit_utils_data" ) self.pretrained_model_path = cache_dir self.init_data_path = os.path.join(cache_dir, "init_data.npz") # download model and data if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(PRETRAINED_MODEL_URL, cache_dir, md5=PRETRAINED_MODEL_MD5) get_data_from_cache( PRETRAINED_MODEL_CONFIG_URL, cache_dir, md5=PRETRAINED_MODEL_CONFIG_MD5 ) get_data_from_cache(INIT_DATA, cache_dir, md5=INIT_DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() # prepare input data self.input_image = np.load(self.init_data_path)["arr_0"] @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_vit_loader_with_data_tensor_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = ViTLoaderHuggerFace( model=libai.models.VisionTransformer, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, ) model = load_func.load() model.eval() input_image = flow.tensor( self.input_image.tolist(), dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] self.assertTrue( np.allclose(np.array(3.1374), prediction_scores.sum().data.numpy(), 1e-4, 1e-4) ) @flow.unittest.skip_unless_1n4d() def test_vit_loader_with_data_tensor_pipeline_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = ViTLoaderHuggerFace( model=libai.models.VisionTransformer, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, ) model = load_func.load() model.eval() input_image = flow.tensor( self.input_image, dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] self.assertTrue( np.allclose(np.array(3.1374), prediction_scores.sum().data.numpy(), 1e-4, 1e-4) ) @flow.unittest.skip_unless_1n4d() def test_vit_loader_with_data_tensor_parallel_backward(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = ViTLoaderHuggerFace( model=libai.models.VisionTransformer, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, drop_rate=0, attn_drop_rate=0, drop_path_rate=0, ) model = load_func.load() input_image = flow.tensor( self.input_image, dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] loss = prediction_scores.sum() loss.backward() self.assertTrue(np.allclose(-173459.77, model.head.weight.grad.sum().numpy(), 1e-3)) @flow.unittest.skip_unless_1n4d() def test_vit_loader_with_data_tensor_pipeline_parallel_backward(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = ViTLoaderHuggerFace( model=libai.models.VisionTransformer, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, drop_rate=0, attn_drop_rate=0, drop_path_rate=0, ) model = load_func.load() input_image = flow.tensor( self.input_image, dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] loss = prediction_scores.sum() loss.backward() self.assertTrue(np.allclose(-173459.77, model.head.weight.grad.sum().numpy(), 1e-3)) if __name__ == "__main__": unittest.main()
7,553
33.493151
151
py
libai
libai-main/tests/model_loader/test_bert_loader.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import numpy as np import oneflow as flow import oneflow.unittest from omegaconf import DictConfig import libai from configs.common.models.bert import cfg as libai_cfg from libai.models.utils import BertLoaderHuggerFace from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger PRETRAINED_MODEL_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/bert_utils/pytorch_model.bin" # noqa PRETRAINED_MODEL_CONFIG_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/bert_utils/config.json" # noqa PRETRAINED_MODEL_MD5 = "ea97b42698d3b5f6d8e8011eba3d1611" PRETRAINED_MODEL_CONFIG_MD5 = "0939b914fc32135f6c12d8ef281dbd7a" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_bert_utils") setup_logger(distributed_rank=dist.get_rank()) class TestBertLoder(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join( os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "bert_utils_data" ) self.pretrained_model_path = cache_dir # prepare dataset if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(PRETRAINED_MODEL_URL, cache_dir, md5=PRETRAINED_MODEL_MD5) get_data_from_cache( PRETRAINED_MODEL_CONFIG_URL, cache_dir, md5=PRETRAINED_MODEL_CONFIG_MD5 ) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() # prepare input data self.input_ids = [ [101, 2009, 1005, 1055, 2986, 2651, 1012, 102], [101, 2028, 12314, 3377, 102, 0, 0, 0], [101, 2064, 2017, 3305, 2009, 102, 0, 0], ] self.mask = [[1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1]] @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_bert_loader_with_data_tensor_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = BertLoaderHuggerFace( model=libai.models.BertModel, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=False, apply_query_key_layer_scaling=False, apply_residual_post_layernorm=True, amp_enabled=False, ) model = load_func.load() model.eval() input_ids = flow.tensor( self.input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.vocab_embeddings.weight.placement, ) mask = flow.tensor( self.mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.vocab_embeddings.weight.placement, ) last_hidden_state, _ = model(input_ids, mask) self.assertTrue( np.allclose(np.array(-214.9335), last_hidden_state.sum().data.numpy(), 1e-4, 1e-4) ) @flow.unittest.skip_unless_1n4d() def test_bert_loader_with_data_tensor_pipeline_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = BertLoaderHuggerFace( model=libai.models.BertModel, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, bias_gelu_fusion=False, bias_dropout_fusion=False, scale_mask_softmax_fusion=False, apply_query_key_layer_scaling=False, apply_residual_post_layernorm=True, amp_enabled=False, ) model = load_func.load() model.eval() input_ids = flow.tensor( self.input_ids, dtype=flow.long, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.vocab_embeddings.weight.placement, ) mask = flow.tensor( self.mask, dtype=flow.bool, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.embeddings.vocab_embeddings.weight.placement, ) last_hidden_state, _ = model(input_ids, mask) self.assertTrue( np.allclose(np.array(-214.9335), last_hidden_state.sum().data.numpy(), 1e-4, 1e-4) ) if __name__ == "__main__": unittest.main()
5,969
34.325444
152
py
libai
libai-main/tests/model_loader/__init__.py
0
0
0
py
libai
libai-main/tests/model_loader/test_swinv2_loader.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil import unittest import numpy as np import oneflow as flow import oneflow.unittest from omegaconf import DictConfig import libai from configs.common.models.swinv2.swinv2_tiny_patch4_window8_256 import cfg as libai_cfg from libai.models.utils import SwinV2LoaderHuggerFace from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache from libai.utils.logger import setup_logger PRETRAINED_MODEL_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/swinv2_utils/pytorch_model.bin" # noqa PRETRAINED_MODEL_CONFIG_URL = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/swinv2_utils/config.json" # noqa INIT_DATA = "http://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/model_utils_test/swinv2_utils/init_data.npz" # noqa PRETRAINED_MODEL_MD5 = "40f085f8916974dcb5d86fc6e03aa0df" PRETRAINED_MODEL_CONFIG_MD5 = "2d3874d58f3d5684f51f70ca29a7de9f" INIT_DATA_MD5 = "c19b2ad8afe9a708aac9d2a0ff15f7bd" TEST_OUTPUT = os.path.join(os.getenv("TEST_OUTPUT", "output_unittest"), "test_swinv2_utils") setup_logger(distributed_rank=dist.get_rank()) class TestSwinV2Loder(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join( os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "swinv2_utils_data" ) self.pretrained_model_path = cache_dir self.init_data_path = os.path.join(cache_dir, "init_data.npz") # download model and data if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(PRETRAINED_MODEL_URL, cache_dir, md5=PRETRAINED_MODEL_MD5) get_data_from_cache( PRETRAINED_MODEL_CONFIG_URL, cache_dir, md5=PRETRAINED_MODEL_CONFIG_MD5 ) get_data_from_cache(INIT_DATA, cache_dir, md5=INIT_DATA_MD5) os.makedirs(TEST_OUTPUT, exist_ok=True) dist.synchronize() # prepare input data self.input_image = np.load(self.init_data_path)["arr_0"] @classmethod def tearDownClass(cls) -> None: if os.path.isdir(TEST_OUTPUT) and dist.get_local_rank() == 0: shutil.rmtree(TEST_OUTPUT) @flow.unittest.skip_unless_1n4d() def test_swinv2_loader_with_data_tensor_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = SwinV2LoaderHuggerFace( model=libai.models.SwinTransformerV2, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, ) model = load_func.load() model.eval() input_image = flow.tensor( self.input_image.tolist(), dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] self.assertTrue( np.allclose(np.array(221.7827), prediction_scores.sum().data.numpy(), 1e-4, 1e-4) ) @flow.unittest.skip_unless_1n4d() def test_swinv2_loader_with_data_tensor_pipeline_parallel(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = SwinV2LoaderHuggerFace( model=libai.models.SwinTransformerV2, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, ) model = load_func.load() model.eval() input_image = flow.tensor( self.input_image, dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] self.assertTrue( np.allclose(np.array(221.7827), prediction_scores.sum().data.numpy(), 1e-4, 1e-4) ) @flow.unittest.skip_unless_1n4d() def test_swinv2_loader_with_data_tensor_parallel_backward(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = SwinV2LoaderHuggerFace( model=libai.models.SwinTransformerV2, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, drop_rate=0, drop_path_rate=0, ) model = load_func.load() input_image = flow.tensor( self.input_image.tolist(), dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] loss = prediction_scores.sum() loss.backward() self.assertTrue(np.allclose(373520.47, model.head.weight.grad.sum().numpy(), 1e-3)) self.assertTrue( np.allclose(259.379, model.patch_embed.norm.weight.grad.sum().numpy(), 1e-3) ) @flow.unittest.skip_unless_1n4d() def test_swinv2_loader_with_data_tensor_pipeline_parallel_backward(self): # set distributed config dist_cfg = DictConfig( dict( data_parallel_size=2, tensor_parallel_size=1, pipeline_parallel_size=2, pipeline_num_layers=12, ) ) dist.setup_dist_util(dist_cfg) # load model load_func = SwinV2LoaderHuggerFace( model=libai.models.SwinTransformerV2, libai_cfg=libai_cfg, pretrained_model_path=self.pretrained_model_path, drop_rate=0, drop_path_rate=0, ) model = load_func.load() input_image = flow.tensor( self.input_image.tolist(), dtype=flow.float32, sbp=dist.get_nd_sbp([flow.sbp.broadcast, flow.sbp.broadcast]), placement=model.patch_embed.proj.weight.placement, ) prediction_scores = model(input_image)["prediction_scores"] loss = prediction_scores.sum() loss.backward() self.assertTrue(np.allclose(373520.47, model.head.weight.grad.sum().numpy(), 1e-3)) self.assertTrue( np.allclose(259.379, model.patch_embed.norm.weight.grad.sum().numpy(), 1e-3) ) if __name__ == "__main__": unittest.main()
7,819
34.067265
154
py
libai
libai-main/tests/layers/test_trainer_model.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow from oneflow import nn class demo_model(nn.Module): def __init__(self, input_dim=512, out_dim=64): super().__init__() self.linear1 = nn.Linear(input_dim, out_dim) self.linear2 = nn.Linear(out_dim, out_dim) def forward(self, x): x = self.linear1(x) x = self.linear2(x) loss = self.get_loss(x) return loss def get_loss(self, x): return x.sum() def build_model(cfg): model = demo_model() placement = flow.env.all_device_placement("cuda") model = model.to_global(placement=placement, sbp=flow.sbp.broadcast) return model def build_graph(cfg, model, optimizer, lr_scheduler, fp16=False): class GraphModel(nn.Graph): def __init__(self, model, optimizer, lr_scheduler, fp16=False): super().__init__() self.model = model self.add_optimizer(optimizer, lr_sch=lr_scheduler) self.config.allow_fuse_add_to_output(True) self.config.allow_fuse_model_update_ops(True) if fp16: self.config.enable_amp(True) grad_scaler = flow.amp.GradScaler( init_scale=2 ** 30, growth_factor=2.0, backoff_factor=0.5, growth_interval=2000, ) self.set_grad_scaler(grad_scaler) def build(self, x): loss = self.model(x) loss.backward() return loss if optimizer: return GraphModel(model, optimizer, lr_scheduler, fp16=fp16) else: return None
2,247
31.114286
74
py
libai
libai-main/tests/layers/test_evaluator_model.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import oneflow as flow from oneflow import nn from libai.models.utils.graph_base import GraphBase class demo_model(nn.Module): def __init__(self, input_dim=512, out_dim=3): super().__init__() self.linear1 = nn.Linear(input_dim, input_dim // 2) self.linear2 = nn.Linear(input_dim // 2, out_dim) def forward(self, x, label=None): x = x.to(dtype=flow.float32) x = self.linear1(x) x = self.linear2(x) if label is None: return x loss = self.get_loss(x) return loss def get_loss(self, x): return x.sum() def build_model(cfg): model = demo_model() placement = flow.env.all_device_placement("cuda") model = model.to_global(placement=placement, sbp=flow.sbp.broadcast) return model class GraphModel(GraphBase): def build(self, x, label=None): if self.is_train: loss = self.model(x, label) loss.backward() return loss else: return self.model(x)
1,653
28.535714
74
py
libai
libai-main/tests/layers/test_linear.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import unittest import numpy as np import oneflow as flow import oneflow.unittest from omegaconf import DictConfig from oneflow import nn from libai.layers import Linear from libai.utils import distributed as dist class TestLinear(flow.unittest.TestCase): @unittest.skipIf(not flow.cuda.is_available(), "only test gpu cases") @flow.unittest.skip_unless_1n1d() def test_nn_linear(self): dist.setup_dist_util( DictConfig( dict( data_parallel_size=1, tensor_parallel_size=1, pipeline_parallel_size=1, ) ) ) inputs = flow.rand(8, 8, sbp=flow.sbp.broadcast, placement=dist.get_layer_placement(0)) weight = flow.rand(4, 8, sbp=flow.sbp.broadcast, placement=dist.get_layer_placement(0)) bias = flow.rand(4, sbp=flow.sbp.broadcast, placement=dist.get_layer_placement(0)) nn_linear = nn.Linear(8, 4).to("cuda") nn_linear.weight.data.copy_(dist.ttol(weight).to("cuda")) nn_linear.bias.data.copy_(dist.ttol(bias).to("cuda")) nn_output = nn_linear(dist.ttol(inputs).to("cuda")) libai_linear = Linear(8, 4) libai_linear.weight.data.copy_(weight) libai_linear.bias.data.copy_(bias) libai_output = libai_linear(inputs) self.assertTrue(np.allclose(nn_output.cpu().numpy(), dist.tton(libai_output), 1e-7, 1e-7)) @flow.unittest.skip_unless_1n2d() def test_col_parallel_linear(self): dist.setup_dist_util( DictConfig( dict( data_parallel_size=1, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) ) inputs = flow.rand(8, 8, sbp=flow.sbp.broadcast, placement=dist.get_layer_placement(0)) weight = flow.rand(4, 8, sbp=flow.sbp.split(0), placement=dist.get_layer_placement(0)) bias = flow.rand(4, sbp=flow.sbp.split(0), placement=dist.get_layer_placement(0)) nn_linear = nn.Linear(8, 4).to("cuda") nn_linear.weight.data.copy_(dist.ttol(weight).to("cuda")) nn_linear.bias.data.copy_(dist.ttol(bias).to("cuda")) nn_output = nn_linear(dist.ttol(inputs).to("cuda")) libai_linear = Linear(8, 4, parallel="col") libai_linear.weight.data.copy_(weight) libai_linear.bias.data.copy_(bias) libai_output = libai_linear(inputs) self.assertTrue(np.allclose(nn_output.cpu().numpy(), dist.tton(libai_output), 1e-7, 1e-7)) @flow.unittest.skip_unless_1n2d() def test_row_parallel_linear(self): dist.setup_dist_util( DictConfig( dict( data_parallel_size=1, tensor_parallel_size=2, pipeline_parallel_size=1, ) ) ) inputs = flow.rand(8, 8, sbp=flow.sbp.broadcast, placement=dist.get_layer_placement(0)) weight = flow.rand(4, 8, sbp=flow.sbp.split(1), placement=dist.get_layer_placement(0)) bias = flow.rand(4, sbp=flow.sbp.broadcast, placement=dist.get_layer_placement(0)) # move local tensor to cuda nn_linear = nn.Linear(8, 4).to("cuda") nn_linear.weight.data.copy_(dist.ttol(weight).to("cuda")) nn_linear.bias.data.copy_(dist.ttol(bias).to("cuda")) nn_output = nn_linear(dist.ttol(inputs).to("cuda")) libai_linear = Linear(8, 4, parallel="row") libai_linear.weight.data.copy_(weight) libai_linear.bias.data.copy_(bias) libai_output = libai_linear(inputs) self.assertTrue(np.allclose(nn_output.cpu().numpy(), dist.tton(libai_output), 1e-7, 1e-7))
4,370
37.008696
98
py
libai
libai-main/tests/layers/__init__.py
0
0
0
py
libai
libai-main/tests/data/test_sampler.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # import itertools # import unittest # import oneflow.utils.data as flowdata # from libai.data.samplers import CyclicSampler, SingleRoundSampler # class TestCyclicSampler(unittest.TestCase): # def test_cyclic_sampler_iterable(self): # sampler = CyclicSampler( # list(range(100)), # micro_batch_size=4, # shuffle=True, # consumed_samples=0, # seed=123, # ) # output_iter = itertools.islice(sampler, 25) # iteration=100/4=25 # sample_output = list() # for batch in output_iter: # sample_output.extend(batch) # self.assertEqual(set(sample_output), set(range(100))) # data_sampler = CyclicSampler( # list(range(100)), # micro_batch_size=4, # shuffle=True, # consumed_samples=0, # seed=123, # ) # data_loader = flowdata.DataLoader( # list(range(100)), batch_sampler=data_sampler, num_workers=0, collate_fn=lambda x: x # ) # data_loader_iter = itertools.islice(data_loader, 25) # output = list() # for data in data_loader_iter: # output.extend(data) # self.assertEqual(output, sample_output) # def test_cyclic_sampler_seed(self): # sampler = CyclicSampler( # list(range(100)), # micro_batch_size=4, # shuffle=True, # seed=123, # ) # data = list(itertools.islice(sampler, 65)) # sampler = CyclicSampler( # list(range(100)), # micro_batch_size=4, # shuffle=True, # seed=123, # ) # data2 = list(itertools.islice(sampler, 65)) # self.assertEqual(data, data2) # def test_cyclic_sampler_resume(self): # # Single rank # sampler = CyclicSampler( # list(range(10)), # micro_batch_size=4, # shuffle=True, # seed=123, # ) # all_output = list(itertools.islice(sampler, 50)) # iteration 50 times # sampler = CyclicSampler( # list(range(10)), # micro_batch_size=4, # shuffle=True, # seed=123, # consumed_samples=4 * 11, # consumed 11 iters # ) # resume_output = list(itertools.islice(sampler, 39)) # self.assertEqual(all_output[11:], resume_output) # def test_cyclic_sampler_resume_multi_rank(self): # # Multiple ranks # sampler_rank0 = CyclicSampler( # list(range(10)), # micro_batch_size=4, # shuffle=True, # seed=123, # data_parallel_rank=0, # data_parallel_size=2, # ) # sampler_rank1 = CyclicSampler( # list(range(10)), # micro_batch_size=4, # shuffle=True, # seed=123, # data_parallel_rank=1, # data_parallel_size=2, # ) # all_output_rank0 = list(itertools.islice(sampler_rank0, 50)) # iteration 50 times # all_output_rank1 = list(itertools.islice(sampler_rank1, 50)) # iteration 50 times # sampler_rank0 = CyclicSampler( # list(range(10)), # micro_batch_size=4, # shuffle=True, # seed=123, # data_parallel_rank=0, # data_parallel_size=2, # consumed_samples=4 * 11, # consumed 11 iters # ) # sampler_rank1 = CyclicSampler( # list(range(10)), # micro_batch_size=4, # shuffle=True, # seed=123, # data_parallel_rank=1, # data_parallel_size=2, # consumed_samples=4 * 11, # consumed 11 iters # ) # resume_output_rank0 = list(itertools.islice(sampler_rank0, 39)) # resume_output_rank1 = list(itertools.islice(sampler_rank1, 39)) # self.assertEqual(all_output_rank0[11:], resume_output_rank0) # self.assertEqual(all_output_rank1[11:], resume_output_rank1) # class TestSingleRoundSampler(unittest.TestCase): # def test_single_sampler_iterable(self): # sampler = SingleRoundSampler( # list(range(100)), # micro_batch_size=4, # shuffle=False, # ) # output_iter = itertools.islice(sampler, 30) # exceed iteration number # sample_output = list() # for batch in output_iter: # sample_output.extend(batch) # self.assertEqual(sample_output, list(range(100))) # def test_single_sampler_multi_rank(self): # sampler_rank0 = SingleRoundSampler( # list(range(101)), # micro_batch_size=4, # shuffle=False, # data_parallel_rank=0, # data_parallel_size=2, # ) # sampler_rank1 = SingleRoundSampler( # list(range(101)), # micro_batch_size=4, # shuffle=False, # data_parallel_rank=1, # data_parallel_size=2, # ) # output_iter_rank0 = itertools.islice(sampler_rank0, 30) # sample_output_rank0 = list() # for batch in output_iter_rank0: # sample_output_rank0.extend(batch) # output_iter_rank1 = itertools.islice(sampler_rank1, 30) # sample_output_rank1 = list() # for batch in output_iter_rank1: # sample_output_rank1.extend(batch) # # Padding 0 if it's not enough for a batch, otherwise `to_global` # # will raise errors for imbalanced data shape in different ranks # self.assertEqual(sample_output_rank0, list(range(51))) # self.assertEqual(sample_output_rank1, list(range(51, 101)) + [0]) # if __name__ == "__main__": # unittest.main()
6,458
32.293814
97
py
libai
libai-main/tests/data/__init__.py
0
0
0
py
libai
libai-main/tests/inference/test_text_generation.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import unittest import numpy as np import oneflow as flow import oneflow.unittest from libai.inference.text_generation import TextGenerationPipeline from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache VOCAB_URL = "https://oneflow-static.oss-cn-beijing.aliyuncs.com/ci-files/dataset/libai/bert_dataset/bert-base-chinese-vocab.txt" # noqa VOCAB_MD5 = "65ac8a72466e859cd3c6b279ed8e532a" class TestTextGenerationPipeline(flow.unittest.TestCase): def setUp(self) -> None: self.texts = ["cat ", "you ", "dog ", "dragon ", "牛 ", "羊 "] cache_dir = os.path.join(os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "bert_data") # prepare tokenizer if dist.get_local_rank() == 0: # download tokenzier vocab on main process of each node get_data_from_cache(VOCAB_URL, cache_dir, md5=VOCAB_MD5) @unittest.skipIf(not flow.cuda.is_available(), "only test gpu cases") @flow.unittest.skip_unless_1n4d() def test_pipeline_with_tensor_parallel(self): self.pipeline = TextGenerationPipeline("configs/t5_large_pretrain.py", 1, 4, 1) for _ in range(5): text = list(np.random.randint(0, 5, 10)) text = "".join([self.texts[i] for i in text]) dict1 = self.pipeline( text, use_cache=False, max_generate_length=15, return_type="new_text" ) dict2 = self.pipeline( text, use_cache=True, max_generate_length=15, return_type="new_text" ) if dist.is_main_process(): assert dict1["generated_text"] == dict2["generated_text"] @unittest.skipIf(not flow.cuda.is_available(), "only test gpu cases") @flow.unittest.skip_unless_1n4d() def test_pipeline_with_pipeline_parallel(self): self.pipeline = TextGenerationPipeline("configs/t5_large_pretrain.py", 1, 1, 4) for _ in range(5): text = list(np.random.randint(0, 5, 10)) text = "".join([self.texts[i] for i in text]) dict1 = self.pipeline( text, use_cache=False, max_generate_length=15, return_type="new_text" ) dict2 = self.pipeline( text, use_cache=True, max_generate_length=15, return_type="new_text" ) if dist.is_main_process(): assert dict1["generated_text"] == dict2["generated_text"] @unittest.skipIf(not flow.cuda.is_available(), "only test gpu cases") @flow.unittest.skip_unless_1n4d() def test_pipeline_with_tensor_pipeline_parallel(self): self.pipeline = TextGenerationPipeline("configs/t5_large_pretrain.py", 1, 2, 2) for _ in range(5): text = list(np.random.randint(0, 5, 10)) text = "".join([self.texts[i] for i in text]) dict1 = self.pipeline( text, use_cache=False, max_generate_length=15, return_type="new_text" ) dict2 = self.pipeline( text, use_cache=True, max_generate_length=15, return_type="new_text" ) if dist.is_main_process(): assert dict1["generated_text"] == dict2["generated_text"] if __name__ == "__main__": unittest.main()
3,903
40.094737
136
py
libai
libai-main/tests/inference/test_image_classification.py
# coding=utf-8 # Copyright 2021 The OneFlow Authors. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import unittest import zipfile import numpy as np import oneflow as flow import oneflow.unittest from libai.inference.image_classification import ImageClassificationPipeline from libai.utils import distributed as dist from libai.utils.file_utils import get_data_from_cache IMAGE_URL = "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/LiBai/Inference/ILSVRC2012_val_00000293.JPEG" # noqa CONFIG_URL = "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/LiBai/ImageNet/vit_small_patch16_224/config.yaml" # noqa MODEL_URL = "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/LiBai/ImageNet/vit_small_patch16_224/model_best.zip" # noqa IMAGE_MD5 = "65ac8a72466e859cd3c6b279ed8e532a" CONFIG_MD5 = "4cf8e662d76f855f4d99ce7129050e79" MODEL_MD5 = "2bfc9cb7df5739d1a1d11db97f54d93f" def _legacy_zip_load(filename, model_dir): # Note: extractall() defaults to overwrite file if exists. No need to clean up beforehand. # We deliberately don't handle tarfile here since our legacy serialization format was in tar. with zipfile.ZipFile(filename) as f: members = f.infolist() extracted_name = members[0].filename extracted_file = os.path.join(model_dir, extracted_name) if not os.path.exists(extracted_file): os.mkdir(extracted_file) f.extractall(model_dir) class TestImageClassificationPipeline(flow.unittest.TestCase): def setUp(self) -> None: cache_dir = os.path.join( os.getenv("ONEFLOW_TEST_CACHE_DIR", "./data_test"), "inference_test_data" ) model_path = os.path.join(cache_dir, MODEL_URL.split("/")[-1]) if dist.get_local_rank() == 0: # download dataset on main process of each node get_data_from_cache(IMAGE_URL, cache_dir, md5=IMAGE_MD5) get_data_from_cache(CONFIG_URL, cache_dir, md5=CONFIG_MD5) get_data_from_cache(MODEL_URL, cache_dir, md5=MODEL_MD5) _legacy_zip_load(model_path, os.path.dirname(model_path)) self.image_path = os.path.join(cache_dir, IMAGE_URL.split("/")[-1]) self.config_path = os.path.join(cache_dir, CONFIG_URL.split("/")[-1]) self.model_path = model_path.replace(".zip", "") assert os.path.isdir(self.model_path) @unittest.skipIf(not flow.cuda.is_available(), "only test gpu cases") @flow.unittest.skip_unless_1n4d() def test_smallvitpipeline_with_pipeline_parallel(self): self.pipeline = ImageClassificationPipeline(self.config_path, 1, 1, 4, self.model_path) rst = self.pipeline(self.image_path) if flow.env.get_rank() == 0: self.assertTrue(rst["label"] == "tench, Tinca tinca") self.assertTrue( np.allclose(np.array(0.7100194096565247), np.array(rst["score"]), 1e-4, 1e-4) ) @unittest.skipIf(not flow.cuda.is_available(), "only test gpu cases") @flow.unittest.skip_unless_1n4d() def test_pipeline_with_pipeline_parallel(self): self.pipeline = ImageClassificationPipeline("configs/vit_imagenet.py", 1, 1, 4) self.pipeline(self.image_path) @unittest.skipIf(not flow.cuda.is_available(), "only test gpu cases") @flow.unittest.skip_unless_1n4d() def test_pipeline_with_tensor_parallel(self): pass # TODO: bug occurs when tensor parallel # self.pipeline = ImageClassificationPipeline("configs/vit_imagenet.py", 1, 4, 1) # self.pipeline(self.image_path) if __name__ == "__main__": unittest.main()
4,163
42.831579
134
py
libai
libai-main/tests/inference/__init__.py
0
0
0
py