soumith ganhacks issues

soumith / ganhacks

starter from "How to Train a GAN?" at NIPS2016

11.43k stars 1.66k forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Generator and discriminator schedule

#77 jkdev100 opened 10 months ago
0
GAN training example

#76 wendlerc opened 10 months ago
0
Activation functions for training and testing phases

#75 SITUSITU opened 1 year ago
1
MAE increasing when training!

#74 DISAPPEARED13 opened 1 year ago
0
CycleGAN

#73 Blakey-Gavin opened 2 years ago
0
Tricks to sharpen GAN's synthesis

#72 DISAPPEARED13 opened 2 years ago
0
Fix a typo

#71 Yunyung opened 2 years ago
0
Fix a grammar error

#70 xTayEx opened 2 years ago
0
Dose it need to transfer pretrained discriminator when transfer learning for GAN?

#69 micklexqg opened 3 years ago
1
Reference for trick#16

#68 shaurov2253 opened 3 years ago
0
> @KrnTneja : Thanks for your tricks. Could you provide any code to do it? I also meet the problem of loss D goes to zero

#67 shaurov2253 closed 4 years ago
1
Discriminator loss of WGAN-gp drop sharply to zero

#66 AboMad opened 4 years ago
0
balance loss maybe is good idea?

#65 Oktai15 opened 4 years ago
0
Hacks relevancy in 2020

#64 phillipiv opened 4 years ago
1
Understand #10

#63 fantasy-fish opened 4 years ago
0
Paper / talk associated?

#62 cossio closed 4 years ago
0
A proposed Variational Autoencoder Hack: Focal Loss with greedy pretraining

#61 PsycheShaman opened 4 years ago
0
Evaluate the Discriminator module over different set of images

#60 anamika06jain opened 5 years ago
0
PixelShuffle, ConvTranspose2d + stride will get bad checkerboard

#59 gmt710 opened 5 years ago
2
is the training normal? how is the right train loss curve?

#58 micklexqg opened 5 years ago
0
Unexplainable GAN losses, need help

#57 lennert130994 opened 5 years ago
3
Need help in understanding #1 and #16

#56 miqbal23 opened 5 years ago
2
added WGAN method and spectral norm

#55 miqbal23 opened 5 years ago
0
Question: Any interest in putting together a premium version for sale on SugarKubes?

#54 wrannaman closed 5 years ago
3
label smoothing one-sided?

#53 makeyourownalgorithmicart opened 5 years ago
0
Why "Construct different mini-batches for real and fake, i.e. each mini-batch needs to contain only all real images or all generated images"?

#52 mrgloom opened 5 years ago
3
Regarding sampling image in batches

#51 miqbal23 opened 5 years ago
0
added simple trick

#50 ajayrfhp opened 5 years ago
0
Generator try to maximize, Discriminator try to minimize

#49 Auth0rM0rgan closed 4 years ago
1
Generator ADAM X Discriminator SGD

#48 nuneslu opened 6 years ago
3
G is always really smaller than D

#47 kristosh closed 6 years ago
0
Standard G & D Loss Values & Trends?

#46 James-G-Hill opened 6 years ago
5
Discriminator with accuracy 1, generator fools it perfectly. What's going on?

#45 danielegrattarola opened 6 years ago
5
Discriminator accuracy: 0.5 on average, but collapsed into 0 for negative samples and 1 for positive samples

#44 danielegrattarola closed 6 years ago
2
Using generator and discriminator loss to evaluate GANs.

#43 Sinha-Raunak opened 6 years ago
0
GAN - Multiple iterations for generator training

#42 ahmed-fau closed 6 years ago
0
Label smoothing

#41 knazeri opened 6 years ago
0
How to handle failure mode

#40 solomatov opened 6 years ago
3
Regarding 13: Add noise to inputs...

#39 vmvargas opened 6 years ago
1
Saturation of G_loss and D_loss

#38 rishabhkumar296 opened 6 years ago
0
How hard it is to make a GAN experiment reproducible?

#37 sbmaruf opened 6 years ago
0
Why Discriminator Loss 0 is a failure mode ?

#36 arijitx opened 6 years ago
15
Alternating training vs. DCGAN?

#35 meder411 opened 6 years ago
0
How would one go about training the network with one channel images?

#34 patakk opened 6 years ago
2
Feasible loss simplification?

#33 florian-boehm opened 6 years ago
2
about generator loss

#32 fsalmasri opened 6 years ago
0
There is no source that has empirically demonstrated a positive effect of replay buffer on GAN training

#31 AranKomat closed 6 years ago
1
measuring quality

#30 malie opened 6 years ago
0
for GAN ，why my D loss is increse，and G loss decrease to 0 at the begining

#29 hefeiwangyande opened 6 years ago
10
Sampling from a sphere instead of a cube

#28 mgarbade opened 6 years ago
3