Same resolution output image

[tthg119]

Hi there, can you please check again your model code? I run the test file with your pretrained models and all default args (with and without --upscale_only flag), the output result image is always the same as the input's. It evenly looks blurrier than the input itself, as upscaled and downscaled again I think? I also tried running with different args but it seems the problem is in your model's output. Also tried with different random images, the output is the same as well.

[tthg119]

I debugged your code and saw that you didn't pass the upscale_only argument in all the functions in data.py.

[amanchadha]

Thanks for identifying the issue. Can you open up a merge request with your changes? That would be well appreciated :)

[amanchadha]

Check with -u, just merged some fixes in https://github.com/amanchadha/iSeeBetter/commit/eb54b27cf044fa85f783729291fb393b73eb082b

[spigelli]

That merge was for an unrelated issue. Seems like it's still downscaling before upscaling even with -u. I'm running:

python3 iSeeBetterTest.py -o <mounted directory> -c --data_dir <dir containing file list and dir of images> --file_list <file list name> --upscale_only

Tested with netG_epoch_4_1 and the FullDataSet version of that model. Here's the docker file for the container I'm running everything in for reference: Dockerfile gist

[nigh8w0lf]

That merge was for an unrelated issue. Seems like it's still downscaling before upscaling even with -u. I'm running:

python3 iSeeBetterTest.py -o <mounted directory> -c --data_dir <dir containing file list and dir of images> --file_list <file list name> --upscale_only

Tested with netG_epoch_4_1 and the FullDataSet version of that model. Here's the docker file for the container I'm running everything in for reference: Dockerfile gist

Yes, I'm seeing the same thing https://github.com/amanchadha/iSeeBetter/issues/21