search

znxlwm / pytorch-generative-model-collections

Collection of generative models in Pytorch version.
2.6k stars 548 forks source link

Doubt in ACGAN #13

Closed devraj89 closed 6 years ago

devraj89 commented 6 years ago

Hi

Thanks for your wonderful code. I read the ACGAN paper and your implementation. I have a query, in the schematic diagram you have shown that c is used along-with x in the Discriminator D. However, in the code, you have not used the class information c while training the D and G(z) as you have done in CGAN ?

[1] Is it wrong to use the class information in the discriminator process ? [2] If it is not required then shouldn't the variable y_fill_ removed ? It seems to be redundant in this case.

Also for the C_fake_loss can we use some random labels or is it required to use those labels that have been sampled from the data_loader (like you have done) ?

Thanks in advance !

znxlwm commented 6 years ago

Hi,

[1] I think the diagram of ACGAN is wrong. discriminator process is right. [2] Yes, you are right. [3] I haven't tested C_fake_loss with random labels. Probably, it is possible. However, we do not need to use random labels because we already have label information.

Thanks.

devraj89 commented 6 years ago

Thanks for your quick reply !

R u planning to incorporate variational autoencoders in the future ?

znxlwm commented 6 years ago

I'm afraid that I have no plans yet.