Deepak Pathak

Yes, you can always modify the input code to do that. I only provided automatic masks for the convenience of code release and to avoid having to release masks file.

@maryam089 Can you paste the full log? Where is this pre-trained network from? Was it trained from the context encoder training code? Are the architecture and other details same?

@NerminSalem @maryam089 Sorry I didn't provide this functionality in the training code (I should have!). But it should not be hard to implement if you look at this [file](https://github.com/pathak22/context-encoder/blob/master/demo.lua#L27) and...

Hi @cryptedp ! Yes, Sorry about that. I will have to retrain the train_random.lua and upload the model at some point. I will put on my todo list to run...

@maryam089 can you post some results here?

@maryam089 Also see #22 (the second picture). Are your results worse than those?

I am still on the way to cleaning the full training code and the environment. It should all be released really soon. Thanks for the patience!

Just in case its not anaconda path issue, can you try without anaconda once?

It should work with anaconda. Removing anaconda is just to check if system vs. anaconda library paths are not entangled i.e. incorrectly specified.

Sorry guys, I have never tried running this on windows. I do not have windows machine, so not sure if I could be of help here.