search

LarryJiang134 / Image_manipulation_detection

Paper: CVPR2018, Learning Rich Features for Image Manipulation Detection
MIT License
360 stars 99 forks source link

Pre trained model error #4

Open aradhyamathur opened 5 years ago

aradhyamathur commented 5 years ago

The link to the pretrained model has not been given. On running demo_two_stream.py I receive the following error 

 Traceback (most recent call last):
  File "demo_two_stream.py", line 124, in <module>
    'our server and place them properly?').format(tfmodel + '.meta'))
OSError: default/DIY_dataset/default/vgg16_faster_rcnn_iter_30000.ckpt.meta not found.
zhangqizky commented 5 years ago

the same error with upstairs

uulucky commented 5 years ago

模型下载地址：https://item.taobao.com/item.htm?id=584772614657

XinZhangRadar commented 5 years ago

Because you have to train the model by yourself...

zqdeepbluesky commented 5 years ago

@XinZhangNLPR can you tell me how to train the model by myself?

XinZhangRadar commented 5 years ago

@XinZhangNLPR can you tell me how to train the model by myself? There are some mistakes in the code, I have wrote a blog can help you to fix it, it was wrote by Chinese, you can translate it and have a look. https://blog.csdn.net/weixin_43380510/article/details/88544830

moulicm111 commented 5 years ago

Even after training the weights are not saved ? Can you please tell me what to change ? I trained it for 3800 iterations .

moulicm111 commented 5 years ago

Got the answer the models will saved after 5000 iterations

saugatapaul1010 commented 5 years ago

Hi Xin,

I am trying to train the model by following your blog. However, I could understand only point one. Can you please mention the steps sequentially, i.e. which files to run one after the other for training the file?