Trying to run the code in dcgan_demo.py but the samples are not created in the 'real' folder of training set.

[xuqiantong]

Trying to run the code in dcgan_demo.py but the samples are not created in the 'real' folder of training set.

[youyuge34]

@xuqiantong I have made a PR which fixed several bugs and modified the structure. Details are shown in PR comment.
THX for your paper and codes again.

[youyuge34]

The bugs are easy to be fixed by following:

• If windows, make the codes in demo_dcgan.py run under
  if __name__ == '__main__': to handle with the multi-process errors.
  (or make the parameter num_worker to 0 in all dataloader() functions)
• open metric.py file in the root directory, line 105-116 which should be

    iter = 0
    for i, data in enumerate(dataloader, 0):
        img, _ = data
        for j in range(0, len(img)):

            vutils.save_image(img[j].mul(0.5).add(
                0.5), saveFolder + giveName(iter) + ".png")
            iter += 1
            if iter >= sampleSize:
                break
        if iter >= sampleSize:
            break

the author input redundant tabs to these lines. use shift+tab to delete the tabs ! That's why no real sample is generated in the folder.

[xuqiantong]

Thanks @youyuge34 for pointing out the problem! Would you mind to send out a PR to fix it?

[xuqiantong]

Bug fixed by @youyuge34 in #2. Closing.