algorithm/pytorch-tutorial
1
0
Fork 0
mirror of https://github.com/yunjey/pytorch-tutorial.git synced 2025-07-27 12:03:31 +08:00
Code Issues Packages Projects Releases Wiki Activity

For truncated BPTT, use built-in detach

Browse Source 
For GANs, detach variables of fake images when training discriminator.
This commit is contained in:
Ke Ding
2017-04-23 05:31:20 -07:00
parent 1aab031bd5
commit d38e95c94d
6 changed files with 12 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tutorials/10 - Generative Adversarial Network/main-gpu.py
View File

@ -77,7 +77,7 @@ for epoch in range(200):
noise = Variable(torch.randn(images.size(0), 128)).cuda()
fake_images = generator(noise)
outputs = discriminator(fake_images)
outputs = discriminator(fake_images.detach())
fake_loss = criterion(outputs, fake_labels)
fake_score = outputs
@ -107,4 +107,4 @@ for epoch in range(200):
# Save the Models
torch.save(generator.state_dict(), './generator.pkl')
torch.save(discriminator.state_dict(), './discriminator.pkl')
torch.save(discriminator.state_dict(), './discriminator.pkl')