Ehud Ben-Reuven

the only way I found was to generate a random batch of images and pick just one. You can look at my code at https://github.com/udibr/dcgan_code/blob/master/acro/train_uncond_dcgan.py it includes many generation features...

this notebook has instructions on how to generate single images (with or without selection of better looking images) https://github.com/udibr/dcgan_code/blob/master/acro/one%20cool%20trick%20for%20GAN.ipynb

@dubtor I managed to run your fork on colab using this notebook https://colab.research.google.com/drive/14nY3p9GG-yfzMziySVqs2zZZk5ArXFiY?usp=sharing

I occasionally do get OOM but most of the time not. It could be its because I'm using "Clab Pro" which gives you a better priority on the GPU card...

I just tried again and indeed got OOM I then did "Runtime->Disconnect and delete run time" and re run the notbook and it works

looks like adding the following code at the *very top* of `run_app.py` helps: ``` import os os.environ['PYTORCH_CUDA_ALLOC_CONF'] = "max_split_size_mb:1000" import torch torch.cuda.empty_cache() ```

@goatchurchprime do you want to run and check PR #4 also add documentation I now have to read the paper to see what each number is

All the predicted parameters fall in range with the paper except for alpha which is around 0,5 in my runs and was 0.2 in paper. Although it was 0.5 in...