Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / ajbrock/BigGAN-PyTorch issues and pull requests

#101 - Lower memory requirements on data prep

Pull Request - State: open - Opened by gcervantes8 over 1 year ago

#100 - Training with 96x96 input

Issue - State: open - Opened by rezendegabriel almost 2 years ago - 1 comment

#99 - The model was trained. How to generate the images?

Issue - State: open - Opened by Mogul5306 about 2 years ago

#98 - Problems when sampling

Issue - State: open - Opened by Howtocreateaname about 2 years ago

#97 - Output 80x688 images

Issue - State: open - Opened by turian over 2 years ago

#96 - set cross_replica as true got bad result

Issue - State: closed - Opened by sunnylulu over 2 years ago - 1 comment

#95 - How to train BigGAN for image generation conditioned on text description?

Issue - State: open - Opened by jaygala24 over 2 years ago - 1 comment

#94 - BigGAN-deep pre-trained

Issue - State: open - Opened by shalabimario over 2 years ago

#93 - GPU Server code

Pull Request - State: closed - Opened by BJohnBraddock over 2 years ago - 1 comment

#92 - Trouble training it from scratch

Issue - State: open - Opened by ParthaEth almost 3 years ago

#91 - collapsed samples

Issue - State: open - Opened by fido20160817 almost 3 years ago

#90 - Query about orthogonal regularization implementation

Issue - State: open - Opened by TanYingHao about 3 years ago

#89 - y_ and y variables in training function

Issue - State: open - Opened by Aveturi13 over 3 years ago

#88 - Correct attention layer implementation

Issue - State: closed - Opened by valillon over 3 years ago - 4 comments

#87 - sampling out of memory

Issue - State: open - Opened by jiyuwangbupt over 3 years ago - 1 comment

#86 - Training with ImageNet 64x64

Issue - State: open - Opened by gulperii over 3 years ago - 1 comment

#85 - requirements.txt or library versions

Issue - State: open - Opened by weroks over 3 years ago

#84 - Out of memory

Issue - State: open - Opened by diaodeyi over 3 years ago - 2 comments

#82 - AttributeError: 'Distribution' object has no attribute 'dist_type'

Issue - State: open - Opened by Ciccios96 almost 4 years ago - 2 comments

#81 - fix preactivation in BigGAN-deep

Pull Request - State: open - Opened by unixpickle almost 4 years ago

#80 - How can I improve the quality of generated images?

Issue - State: open - Opened by pengshy almost 4 years ago

#79 - How can I improve the quality of generated images?

Issue - State: closed - Opened by pengshy almost 4 years ago

#77 - Critical: Code report training FID.

Issue - State: closed - Opened by AlexanderMath almost 4 years ago - 4 comments

#76 - Fix MultiEpochSampler __len__() > 0 error

Pull Request - State: open - Opened by JonFreer almost 4 years ago - 1 comment

#75 - Mode collapse?

Issue - State: open - Opened by Hierakonpolis almost 4 years ago - 4 comments

#74 - fix saving best test metrics

Pull Request - State: open - Opened by mshahbazi72 about 4 years ago

#73 - Dataloader fix

Pull Request - State: closed - Opened by dmckee5 about 4 years ago

#72 - Fix BigGAN-Deep generator; Support 512x512 generator

Pull Request - State: open - Opened by shawwn about 4 years ago

#71 - ValueError: __len__() should return >= 0

Issue - State: open - Opened by JanineCHEN about 4 years ago - 1 comment

#70 - IndexError: tuple index out of range

Issue - State: open - Opened by JanineCHEN about 4 years ago - 4 comments

#69 - Add arch[512] in discriminator and fix arch[256]

Pull Request - State: open - Opened by murbard about 4 years ago

#68 - url error while prepareing data

Issue - State: closed - Opened by banianzr over 4 years ago

#67 - use self.momentum in ccbn

Pull Request - State: open - Opened by Separius over 4 years ago

#66 - D loss is zero from the start

Issue - State: closed - Opened by entrpn over 4 years ago - 5 comments

#65 - "sh scripts/utils/prepare_data.sh" is not Working.

Issue - State: closed - Opened by H1R0Y4 over 4 years ago

#64 - cublas runtime error

Issue - State: open - Opened by phymhan over 4 years ago

#63 - Missing C10_inception_moments.npz

Issue - State: open - Opened by ShuangjunLiu over 4 years ago - 4 comments

#62 - sh scripts/utils/prepare_data.sh error

Issue - State: closed - Opened by ShuangjunLiu over 4 years ago - 1 comment

#61 - Welcome to unofficial TPU enabled PyTorch implementation

Issue - State: open - Opened by shizhediao over 4 years ago - 5 comments

#60 - Hinge loss of generator

Issue - State: open - Opened by yfji over 4 years ago - 1 comment

#59 - How much data is enough for the model to work properly

Issue - State: open - Opened by namtp-ai over 4 years ago

#57 - Singular value clamping code?

Issue - State: closed - Opened by johntiger1 over 4 years ago - 2 comments

#56 - Some guidelines for better gpu perfomance.

Issue - State: open - Opened by ysig over 4 years ago - 2 comments

#54 - How to understand the result given by discriminator?

Issue - State: open - Opened by yqtianust almost 5 years ago - 1 comment

#53 - fix the D_ch in scripts

Pull Request - State: open - Opened by yqtianust almost 5 years ago - 1 comment

#52 - Error encountered in parallel training

Issue - State: closed - Opened by KenyonY almost 5 years ago - 2 comments

#51 - where is the pre-trained model of discriminator?

Issue - State: open - Opened by happycaoyue almost 5 years ago - 2 comments

#49 - FID is nan

Issue - State: open - Opened by doantientai about 5 years ago - 4 comments

#48 - Looking forward to offical TPU-version BigGAN code~

Issue - State: open - Opened by AaronAnima about 5 years ago - 1 comment

#47 - How should I conditional generate samples

Issue - State: closed - Opened by bywbilly about 5 years ago - 1 comment

#46 - Places365

Issue - State: open - Opened by janzd about 5 years ago - 1 comment

#45 - Spectral regularization for combating mode collapse

Issue - State: closed - Opened by max-liu-112 about 5 years ago - 1 comment

#44 - Question regarding the MultiEpochSampler

Issue - State: open - Opened by zxhuang97 about 5 years ago - 1 comment

#43 - fine-tunning

Issue - State: open - Opened by yaxingwang about 5 years ago - 14 comments

#42 - Why drop_last of DataLoader is disabled

Issue - State: closed - Opened by RuiLiFeng about 5 years ago - 3 comments

#41 - May I ask how to resize ImageNet data to 128 by 128?

Issue - State: closed - Opened by UBCDingXin about 5 years ago - 1 comment

#40 - training at 512x512

Issue - State: open - Opened by emanueledelsozzo about 5 years ago - 30 comments

#39 - Running out of memory

Issue - State: closed - Opened by ellismarte over 5 years ago - 8 comments

#38 - No space left on device

Issue - State: closed - Opened by ellismarte over 5 years ago - 2 comments

#37 - Remove duplicated code

Pull Request - State: closed - Opened by jeffling over 5 years ago - 1 comment

#36 - fix ImageFolder init bug with load_in_mem

Pull Request - State: closed - Opened by IShengFang over 5 years ago - 1 comment

#35 - Output images have holes in them

Issue - State: closed - Opened by christegho over 5 years ago - 3 comments

#34 - groupnorm function in layers.py

Issue - State: open - Opened by songyh10 over 5 years ago - 1 comment

#33 - unsupervised GAN without class as conditional input?

Issue - State: closed - Opened by songyh10 over 5 years ago - 2 comments

#32 - A mismatch problem on fine-tune

Issue - State: closed - Opened by chensln over 5 years ago - 3 comments

#31 - A GPU VRAM is 6. Can it be trained?

Issue - State: closed - Opened by LIMr1209 over 5 years ago - 1 comment

#30 - Use my own dataset

Issue - State: closed - Opened by LIMr1209 over 5 years ago - 3 comments

#29 - Using grayscale input images instead of RGB.

Issue - State: closed - Opened by kseuro over 5 years ago - 5 comments

#28 - Can this do conditional image generation?

Issue - State: closed - Opened by LIMr1209 over 5 years ago - 2 comments

#27 - OSError

Issue - State: closed - Opened by hanshan1 over 5 years ago - 1 comment

#26 - Resuming training from given 100k checkpoint collapses

Issue - State: closed - Opened by isaac-dunn over 5 years ago - 1 comment

#25 - Any reason you don't zero optimizer between Discriminator training steps?

Issue - State: closed - Opened by monney over 5 years ago - 2 comments

#24 - Sync Batchnorm

Issue - State: closed - Opened by dougsouza over 5 years ago - 1 comment

#23 - Training results(IS and FID) are not good as yours with same training process

Issue - State: open - Opened by qilimk over 5 years ago - 14 comments

#22 - Are SAGAN and SNGAN scripts tested?

Issue - State: closed - Opened by zoharli over 5 years ago - 1 comment

#21 - How to test on only one GPU?

Issue - State: closed - Opened by VoiceBeer over 5 years ago - 7 comments

#20 - Why channel drop?

Issue - State: closed - Opened by apple2373 over 5 years ago - 1 comment

#19 - Performance on small datasets?

Issue - State: closed - Opened by addiu over 5 years ago - 1 comment

#18 - Do you have any plan to release the Discriminator weights for 256 and 512 models?

Issue - State: closed - Opened by qilimk over 5 years ago - 3 comments

#17 - Any difference between Attention and SAGAN?

Issue - State: closed - Opened by jayzhan211 over 5 years ago - 2 comments

#16 - z_dim for 512x512 model

Issue - State: closed - Opened by ssnl over 5 years ago - 2 comments

#15 - Runtime Error when saving model

Issue - State: closed - Opened by lesnikow over 5 years ago - 6 comments

#14 - Question about truncation trick in Big GAN.

Issue - State: closed - Opened by ShenYujun over 5 years ago - 1 comment

#13 - MemoryError when calculate IS

Issue - State: closed - Opened by limbo0000 over 5 years ago - 1 comment

#12 - correctly update dataset_root when base_root is present

Pull Request - State: closed - Opened by kampta over 5 years ago - 3 comments

#11 - Ignore conversion default save dirs and __pycache__

Pull Request - State: closed - Opened by ssnl over 5 years ago - 1 comment

#10 - Plans to support converting -deep checkpoints?

Issue - State: closed - Opened by ssnl over 5 years ago - 1 comment

#9 - Do you have any plan to release the 256 and 512 pre-trained models?

Issue - State: closed - Opened by qilimk over 5 years ago - 2 comments

#8 - Minimal working example for sampling from pre-trained BigGAN?

Issue - State: closed - Opened by qilimk over 5 years ago - 4 comments

#7 - Should not `import model`

Issue - State: closed - Opened by ssnl over 5 years ago - 1 comment

#6 - how should i test?

Issue - State: closed - Opened by zhangbaijin over 5 years ago - 1 comment

#5 - Fix __all__ to export that class that is defined

Pull Request - State: closed - Opened by cclauss over 5 years ago

#4 - Use ==/!= to compare str, bytes, and int literals

Pull Request - State: closed - Opened by cclauss over 5 years ago - 1 comment