bernhard2202 / improved-video-gan Goto Github PK

Hey I am very interested in GANs and how they can create images and videos.
However I am totally not a coder. I know how to install packages with apt-get and pip, get most dependencies installed but that's about it.
It would be really cool if you could include some more instruction on how to use your code. Especially creating the short videos from still images.
I have a virtualenv with py 2.7 and tensorflow 1.2.1 but this it does not work.
When running improved_video_gan_future.py I get this

~/improved-video-gan/model$ python improved_video_gan_future.py 
Traceback (most recent call last):
  File "improved_video_gan_future.py", line 6, in <module>
    from utils.layers import conv2d, conv3d_transpose, dis_block, linear
ImportError: No module named layers

This is my pip list

~/improved-video-gan/model$ pip list
Package           Version  
----------------- ---------
absl-py           0.2.2    
astor             0.6.2    
backports.weakref 1.0rc1   
Bashutils         0.0.4    
bleach            1.5.0    
enum34            1.1.6    
funcsigs          1.0.2    
futures           3.2.0    
gast              0.2.0    
grpcio            1.13.0   
html5lib          0.9999999
layers            0.1.5    
Markdown          2.6.11   
mock              2.0.0    
numpy             1.14.5   
pbr               4.0.4    
Pillow            5.2.0    
pip               10.0.1   
pkg-resources     0.0.0    
protobuf          3.6.0    
PyYAML            3.12     
setuptools        39.2.0   
six               1.11.0   
tensorflow        1.2.1    
termcolor         1.1.0    
utils             0.9.0    
Werkzeug          0.14.1   
wheel             0.31.1   

sorry for bother u.but when i run the inpaint model i get the issue:

NotFoundError (see above for traceback): Key g_/g_f_bn0/beta/Adam not found in checkpoint
[[Node: save/RestoreV2_80 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/cpu:0"](_arg_save/Const_0_0, save/RestoreV2_80/tensor
localhost/replica:0/task:0/cpu:0"](_arg_save/Const_0_0, save/RestoreV2_80/tensor_names, save/RestoreV2_80/shape_and_slices)]]
[[Node: save/RestoreV2_89/_265 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_506_save/Restore V2_89", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/gpu:0" ]]

@bernhard2202 @d-acharya @zzhiwu
I am very interested in your work, but when I run the program, there is a bug. The bug is located at utils.py->def convert_img(). I want to know what the purpose of this function is?
In addition, the content of the bug is List argument'values' to'ConcatV2' Op with length 0 shorter than minimum length 2. How can I fix this bug?
thank you very much!

I am trying to reproduce the results of the future prediction using the same exact code provided. However, I reached step 1200 but with no successful results. Please advise.

Do you have a pre-trained model downloadable?