Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / soumith/ganhacks issues and pull requests
#78 - Update README.md, fix typo(s)
Pull Request -
State: open - Opened by BobConanDev 3 months ago
#77 - Generator and discriminator schedule
Issue -
State: open - Opened by jkdev100 about 1 year ago
#76 - GAN training example
Issue -
State: open - Opened by wendlerc over 1 year ago
#75 - Activation functions for training and testing phases
Issue -
State: open - Opened by SITUSITU about 2 years ago
- 1 comment
#74 - MAE increasing when training!
Issue -
State: open - Opened by DISAPPEARED13 over 2 years ago
#73 - CycleGAN
Issue -
State: open - Opened by Blakey-Gavin over 2 years ago
#72 - Tricks to sharpen GAN's synthesis
Issue -
State: open - Opened by DISAPPEARED13 almost 3 years ago
#71 - Fix a typo
Pull Request -
State: open - Opened by Yunyung about 3 years ago
#70 - Fix a grammar error
Pull Request -
State: open - Opened by xTayEx about 3 years ago
#69 - Dose it need to transfer pretrained discriminator when transfer learning for GAN?
Issue -
State: open - Opened by micklexqg almost 4 years ago
- 1 comment
#68 - Reference for trick#16
Issue -
State: open - Opened by shaurov2253 over 4 years ago
#67 - > @KrnTneja : Thanks for your tricks. Could you provide any code to do it? I also meet the problem of loss D goes to zero
Issue -
State: closed - Opened by shaurov2253 over 4 years ago
- 1 comment
#66 - Discriminator loss of WGAN-gp drop sharply to zero
Issue -
State: open - Opened by AboMad over 4 years ago
#65 - balance loss maybe is good idea?
Issue -
State: open - Opened by Oktai15 almost 5 years ago
#64 - Hacks relevancy in 2020
Issue -
State: open - Opened by phillipiv almost 5 years ago
- 1 comment
#63 - Understand #10
Issue -
State: open - Opened by fantasy-fish almost 5 years ago
#62 - Paper / talk associated?
Issue -
State: closed - Opened by cossio about 5 years ago
#61 - A proposed Variational Autoencoder Hack: Focal Loss with greedy pretraining
Pull Request -
State: open - Opened by PsycheShaman about 5 years ago
#61 - A proposed Variational Autoencoder Hack: Focal Loss with greedy pretraining
Pull Request -
State: open - Opened by PsycheShaman about 5 years ago
#60 - Evaluate the Discriminator module over different set of images
Issue -
State: open - Opened by anamika06jain over 5 years ago
#59 - PixelShuffle, ConvTranspose2d + stride will get bad checkerboard
Issue -
State: open - Opened by gmt710 over 5 years ago
- 2 comments
#59 - PixelShuffle, ConvTranspose2d + stride will get bad checkerboard
Issue -
State: open - Opened by gmt710 over 5 years ago
- 2 comments
#58 - is the training normal? how is the right train loss curve?
Issue -
State: open - Opened by micklexqg over 5 years ago
#57 - Unexplainable GAN losses, need help
Issue -
State: open - Opened by lennert130994 over 5 years ago
- 3 comments
#56 - Need help in understanding #1 and #16
Issue -
State: open - Opened by miqbal23 over 5 years ago
- 2 comments
#55 - added WGAN method and spectral norm
Pull Request -
State: open - Opened by miqbal23 over 5 years ago
#54 - Question: Any interest in putting together a premium version for sale on SugarKubes?
Issue -
State: closed - Opened by wrannaman almost 6 years ago
- 3 comments
#53 - label smoothing one-sided?
Issue -
State: open - Opened by makeyourownalgorithmicart almost 6 years ago
#52 - 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"?
Issue -
State: open - Opened by mrgloom almost 6 years ago
- 3 comments
#51 - Regarding sampling image in batches
Issue -
State: open - Opened by miqbal23 almost 6 years ago
#50 - added simple trick
Pull Request -
State: open - Opened by ajayrfhp about 6 years ago
#49 - Generator try to maximize, Discriminator try to minimize
Issue -
State: closed - Opened by Auth0rM0rgan over 6 years ago
- 1 comment
#48 - Generator ADAM X Discriminator SGD
Issue -
State: open - Opened by nuneslu over 6 years ago
- 3 comments
#47 - G is always really smaller than D
Issue -
State: closed - Opened by kristosh over 6 years ago
#46 - Standard G & D Loss Values & Trends?
Issue -
State: open - Opened by James-G-Hill over 6 years ago
- 5 comments
#45 - Discriminator with accuracy 1, generator fools it perfectly. What's going on?
Issue -
State: open - Opened by danielegrattarola over 6 years ago
- 5 comments
#44 - Discriminator accuracy: 0.5 on average, but collapsed into 0 for negative samples and 1 for positive samples
Issue -
State: closed - Opened by danielegrattarola over 6 years ago
- 2 comments
#43 - Using generator and discriminator loss to evaluate GANs.
Issue -
State: open - Opened by Sinha-Raunak over 6 years ago
#42 - GAN - Multiple iterations for generator training
Issue -
State: closed - Opened by ahmed-fau over 6 years ago
#41 - Label smoothing
Issue -
State: open - Opened by knazeri over 6 years ago
#40 - How to handle failure mode
Issue -
State: open - Opened by solomatov almost 7 years ago
- 3 comments
#39 - Regarding 13: Add noise to inputs...
Issue -
State: open - Opened by vmvargas almost 7 years ago
- 1 comment
#38 - Saturation of G_loss and D_loss
Issue -
State: open - Opened by rishabhkumar296 almost 7 years ago
#37 - How hard it is to make a GAN experiment reproducible?
Issue -
State: open - Opened by sbmaruf almost 7 years ago
#36 - Why Discriminator Loss 0 is a failure mode ?
Issue -
State: open - Opened by arijitx almost 7 years ago
- 15 comments
#35 - Alternating training vs. DCGAN?
Issue -
State: open - Opened by meder411 about 7 years ago
#34 - How would one go about training the network with one channel images?
Issue -
State: open - Opened by patakk about 7 years ago
- 2 comments
#33 - Feasible loss simplification?
Issue -
State: open - Opened by florian-boehm about 7 years ago
- 2 comments
#32 - about generator loss
Issue -
State: open - Opened by fsalmasri about 7 years ago
#31 - There is no source that has empirically demonstrated a positive effect of replay buffer on GAN training
Issue -
State: closed - Opened by AranKomat about 7 years ago
- 1 comment
#30 - measuring quality
Issue -
State: open - Opened by malie about 7 years ago
#29 - for GAN ,why my D loss is increse,and G loss decrease to 0 at the begining
Issue -
State: open - Opened by hefeiwangyande about 7 years ago
- 10 comments
#28 - Sampling from a sphere instead of a cube
Issue -
State: open - Opened by mgarbade about 7 years ago
- 3 comments
#27 - Update tips for WGANs
Issue -
State: open - Opened by ndahlquist over 7 years ago
- 3 comments
#26 - not working in trick 13: adding gaussian noise to every layer of generator
Issue -
State: open - Opened by danielkaifeng over 7 years ago
- 1 comment
#25 - add G history and historical averaging
Pull Request -
State: closed - Opened by jenkspt over 7 years ago
#24 - Replace BatchNorm with WeightNorm
Issue -
State: open - Opened by andyli over 7 years ago
- 1 comment
#23 - Trick 11: How to schedule thresholds A and B?
Issue -
State: open - Opened by benstaf over 7 years ago
- 2 comments
#22 - Validate a GAN
Issue -
State: open - Opened by dantp-ai over 7 years ago
- 5 comments
#21 - Why are the biases set to zero in the training of GAN?
Issue -
State: open - Opened by hshustc over 7 years ago
#20 - Adam in the generator?
Issue -
State: open - Opened by JasonGUTU over 7 years ago
- 2 comments
#19 - About the generalization of these tricks
Issue -
State: open - Opened by utkarshojha over 7 years ago
#18 - another trick maybe helpful, Could you test on this?
Issue -
State: open - Opened by flybirp almost 8 years ago
#17 - Are the orders ranked?
Issue -
State: open - Opened by jiwoongim about 8 years ago
#16 - When to use labels (12: If you have labels, use them)
Issue -
State: open - Opened by rafaelvalle about 8 years ago
- 2 comments
#15 - Trick 2 explanation
Issue -
State: open - Opened by techmatt about 8 years ago
- 1 comment
#14 - G loss increase, what is this mean?
Issue -
State: open - Opened by wagamamaz about 8 years ago
- 42 comments
#13 - how does trick 6 change ideal output activation/loss
Issue -
State: open - Opened by mjdietzx about 8 years ago
- 1 comment
#12 - Any tricks for text embedding?
Issue -
State: closed - Opened by zsdonghao about 8 years ago
- 3 comments
#11 - Please explain trick 2
Issue -
State: closed - Opened by mjdietzx about 8 years ago
- 3 comments
#10 - Label smoothing should be one-sided?
Issue -
State: closed - Opened by MustafaMustafa about 8 years ago
- 15 comments
#9 - What do you mean by trick 4?
Issue -
State: open - Opened by shuzhangcasia about 8 years ago
- 9 comments
#8 - multiple Discriminator networks
Issue -
State: open - Opened by loliverhennigh about 8 years ago
- 3 comments
#7 - I can't understand this trick
Issue -
State: closed - Opened by zhangqianhui about 8 years ago
- 2 comments
#6 - Have you try more deeper networks?
Issue -
State: open - Opened by zhangqianhui about 8 years ago
- 4 comments
#5 - normal behavior of GAN
Issue -
State: closed - Opened by bobchennan about 8 years ago
- 3 comments
#4 - Added a dropout trick based on https://arxiv.org/pdf/1611.07004v1.pdf
Pull Request -
State: closed - Opened by raminia about 8 years ago
- 1 comment
#3 - Comments or Intuition why these tricks work?
Issue -
State: closed - Opened by spurra about 8 years ago
- 9 comments
#2 - Probable things to investigate when Generator falls behind the Discriminator
Issue -
State: closed - Opened by kmul00 about 8 years ago
- 10 comments
#1 - Update README.md
Pull Request -
State: closed - Opened by zdx3578 about 8 years ago
- 1 comment