text
stringlengths
1
93.6k
# corresponds to doing no classifier free guidance.
do_classifier_free_guidance = guidance_scale > 1.0
# get unconditional embeddings for classifier free guidance
if do_classifier_free_guidance:
max_length = text_input.input_ids.shape[-1]
uncond_input = self.tokenizer(
[""] * batch_size, padding="max_length", max_length=max_length, return_tensors="pt"
)
uncond_embeddings = self.text_encoder(uncond_input.input_ids.to(self.device))[0]
# For classifier free guidance, we need to do two forward passes.
# Here we concatenate the unconditional and text embeddings into a single batch
# to avoid doing two forward passes
text_embeddings = torch.cat([uncond_embeddings, text_embeddings])
# prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
# eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
# eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502
# and should be between [0, 1]
accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
extra_step_kwargs = {}
if accepts_eta:
extra_step_kwargs["eta"] = eta
latents = init_latents
t_start = max(num_inference_steps - init_timestep + offset, 0)
for i, t in tqdm(enumerate(self.scheduler.timesteps[t_start:])):
# expand the latents if we are doing classifier free guidance
latent_model_input = torch.cat([latents] * 2) if do_classifier_free_guidance else latents
# predict the noise residual
noise_pred = self.unet(latent_model_input, t, encoder_hidden_states=text_embeddings)["sample"]
# perform guidance
if do_classifier_free_guidance:
noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
# compute the previous noisy sample x_t -> x_t-1
latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs)["prev_sample"]
# scale and decode the image latents with vae
latents = 1 / 0.18215 * latents
image = self.vae.decode(latents)
image = (image / 2 + 0.5).clamp(0, 1)
image = image.cpu().permute(0, 2, 3, 1).numpy()
# run safety checker
safety_cheker_input = self.feature_extractor(self.numpy_to_pil(image), return_tensors="pt").to(self.device)
#image_, has_nsfw_concept = self.safety_checker(images=image, clip_input=safety_cheker_input.pixel_values)
has_nsfw_concept = []
for hni in range(0, len(image)):
has_nsfw_concept.append(0)
if output_type == "pil":
image = self.numpy_to_pil(image)
return {"sample": image, "nsfw_content_detected": has_nsfw_concept}
# <FILESEP>
#!/usr/local/bin/python
# -*- coding:utf-8 -*-
# @Time : 2019/4/10 10:10 PM
# @Author : Jerry
# @Desc :
# @File : SensitivesHunter.py
import json
import os
import sys
import subprocess
from lib.config import log
from lib.common.basic import getCurrentPath, makeDir, getDomain
from Downloader import DownLoader
from SensitiveFileParser import SensitiveFileParser
import config
class SensitivesHunter():
def __init__(self, url, project_name):
self.start_url = url
self.project_name = project_name
self.crawled_file_links_dict = {}
self.result_dict = {}
self.skip_flag = False # 跳过爬取的标志
def startHunt(self):
self.prepare()
self.tryToSkipCrawled()
if not self.skip_flag:
self.crawlLinks() # 爬取链接
self.parseFileLinks() # 解析爬取到的文件url
for file_type, url_file_list in self.crawled_file_links_dict.items():
downloaded_file_path_dict = self.downloadFile(url_file_list, file_type)
self.detectSensitiveFile(downloaded_file_path_dict, file_type)
self.saveResultFile()
def tryToSkipCrawled(self):
'''