junyanz/BicycleGAN issues and pull requests

#102 - Very Large Images

Issue - State: open - Opened by modtim about 1 year ago

#101 - Hi, please help me.

Issue - State: open - Opened by ZzzTD over 1 year ago

#100 - Why does not large batch size like 128, 256 work well?

Issue - State: open - Opened by bad-meets-joke over 1 year ago

#99 - Regarding Latent Space Interpolation

Issue - State: open - Opened by AvirupJU about 2 years ago - 1 comment

#98 - TypeError: init() got an unexpected keyword argument 'nl_layer'

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

#97 - diversity question

Issue - State: open - Opened by ZhiHong-w over 2 years ago

#96 - Is there <pix2pix+noise> model code that can be directly run？

Issue - State: closed - Opened by qsunyuan about 3 years ago - 1 comment

#95 - How to train on large images?

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

#94 - Regarding Training your Own Images

Issue - State: open - Opened by kuruvilla2087 over 3 years ago - 4 comments

#93 - Incorrect discriminator update for opt.use_same_D

Issue - State: closed - Opened by SanderGielisse almost 4 years ago - 1 comment

#92 - Question: Do you need two separate discriminators?

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

#91 - Question about generating fake_B_random

Issue - State: open - Opened by yeeyangtee about 4 years ago

#90 - Question about Encoder in cLR-GAN

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

#89 - Metric reporting

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

#88 - test_before_push is a great rapid test file, but seems outdated?

Issue - State: open - Opened by NataliaDiaz over 4 years ago - 3 comments

#87 - Compute graph wrong and one question

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

#86 - Test on single images

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

#85 - Not clear in the difference between the two latent spaces predicted

Issue - State: open - Opened by arnabsinha99 over 4 years ago - 5 comments

#84 - Question about conditional_D implementation

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

#83 - Why my LPIPS distance is larger than what your paper say?

Issue - State: open - Opened by Imposingapple over 4 years ago - 3 comments

#82 - Question about PatchGAN perception field

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

#81 - Why we use 'retain_graph' when backwarding loss_G?

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

#80 - Apply BicycleGAN to improve poor line arts

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

#79 - Learn diversity

Issue - State: closed - Opened by remega about 5 years ago - 9 comments

#78 - train_night2day.sh problem

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

#77 - night2day problems

Issue - State: open - Opened by VPRzhou about 5 years ago

#76 - question about the evaluation metrics of diversity

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

#75 - different image size not supported.

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

#74 - Distribution of latent code Z

Issue - State: closed - Opened by hungnv21292 over 5 years ago - 3 comments

#73 - Multi-gpu support---only can run on gpu 0

Issue - State: open - Opened by sgflower66 over 5 years ago - 2 comments

#72 - Using z as an extra label

Issue - State: closed - Opened by yjs0704 over 5 years ago - 2 comments

#71 - D_NLayersMulti implementation correctness

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

#70 - Why cLR-GAN cause a 15% same results?

Issue - State: closed - Opened by xml94 over 5 years ago - 2 comments

#69 - A little question for latent code z.

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

#68 - Ambiguous about one-to-many

Issue - State: closed - Opened by hungnv21292 over 5 years ago - 4 comments

#67 - BicycleGAN on Citycscapes

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

#66 - Expected training time

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

#65 - When to use drop out?

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

#64 - about fake_B_random

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

#63 - Why not freeze netD2

Issue - State: closed - Opened by Ying156209 over 5 years ago - 2 comments

#62 - The result was not very good.

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

#61 - ACB import error

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

#60 - Whether it is right about KL_loss

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

#59 - No such file or directory

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

#58 - not sure if the code is written deliberately about the loss

Issue - State: closed - Opened by xuzhongyou over 5 years ago

#57 - About EOFError: Ran out of input

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

#56 - Hi folks, could you try the latest commit? Hopefully, it can address your issues.

Issue - State: closed - Opened by Shravankp over 5 years ago

#55 - Correct spelling mistake

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

#54 - An Error in train or test `RuntimeError: output with shape [1, 256, 256] doesn't match the broadcast shape [3, 256, 256]`

Issue - State: closed - Opened by shazhongcheng over 5 years ago - 13 comments

#53 - what is this means?

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

#52 - An error in `UnetBlock_with_z`

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

#51 - Issue while running test.py

Issue - State: closed - Opened by a7b23 over 5 years ago - 8 comments

#50 - Differences between released model and training script

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

#49 - LeakyReLU object has no attribute conv

Issue - State: closed - Opened by ic97usop over 5 years ago

#48 - Odd batch sizes

Issue - State: closed - Opened by ic97usop over 5 years ago

#47 - Adding information to BicycleGAN network

Issue - State: closed - Opened by ic97usop almost 6 years ago - 3 comments

#46 - Running BicycleGAN on CPU

Issue - State: closed - Opened by ic97usop almost 6 years ago - 2 comments

#45 - Add missing discriminator identifier

Pull Request - State: closed - Opened by ploth almost 6 years ago

#44 - Remove whitespace

Pull Request - State: closed - Opened by ploth almost 6 years ago

#43 - how was wgan-gp used in the network?

Issue - State: closed - Opened by XiangChen1994 almost 6 years ago - 4 comments

#42 - narrow tensor to zero elements fails

Issue - State: closed - Opened by yhlleo almost 6 years ago - 6 comments

#41 - cuDNN error

Issue - State: closed - Opened by riktimmondal almost 6 years ago - 11 comments
Labels: good first issue

#40 - Fix multi-GPU runtime error with multi-scale netD

Pull Request - State: closed - Opened by cuihaoleo almost 6 years ago - 1 comment

#39 - Installation issue

Issue - State: closed - Opened by MaciekBugajski almost 6 years ago - 1 comment

#38 - About conditioning the discriminator

Issue - State: closed - Opened by r06922019 about 6 years ago - 2 comments

#37 - About disentangled representation of latent code z

Issue - State: closed - Opened by holylone about 6 years ago - 1 comment

#36 - Is there any reason why relu is the default option?

Issue - State: closed - Opened by caffeinism about 6 years ago - 2 comments

#36 - Is there any reason why relu is the default option?

Issue - State: closed - Opened by caffeinism about 6 years ago - 2 comments

#35 - Unexpected pdb debugger trace statement in commit

Issue - State: closed - Opened by lxj616 about 6 years ago - 1 comment

#34 - about multi-gpu support

Issue - State: closed - Opened by pipipopo about 6 years ago - 6 comments

#34 - about multi-gpu support

Issue - State: closed - Opened by pipipopo about 6 years ago - 6 comments

#33 - question about the loss function

Issue - State: closed - Opened by OValery16 over 6 years ago - 2 comments

#32 - About val data

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

#31 - what is the meaning of fake_B_random and fake_B_encoded

Issue - State: closed - Opened by myway0101 over 6 years ago - 2 comments

#30 - Multiple inputs network

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

#29 - Training cLR-GAN or cVAE-GAN

Issue - State: closed - Opened by myway0101 over 6 years ago - 3 comments

#28 - Training on Piano Roll data

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

#27 - Error loading state_dict for G_Unet_add_all

Issue - State: closed - Opened by rafaelvalle over 6 years ago - 1 comment

#26 - How to reproduce the data of LPIPS distance?

Issue - State: closed - Opened by WorkingCC over 6 years ago - 10 comments

#25 - why "self.real_B_random = self.real_B[half_size:]" but not the first half

Issue - State: closed - Opened by civilman628 over 6 years ago - 1 comment

#24 - no odd-length datasets

Pull Request - State: closed - Opened by twak over 6 years ago - 3 comments

#23 - Why use parallel_forward

Issue - State: closed - Opened by HelenMao over 6 years ago - 4 comments

#22 - the database of day2night

Issue - State: closed - Opened by Ruhjkg over 6 years ago - 1 comment

#21 - skip this point data_size = 1

Issue - State: closed - Opened by rharadgithub over 6 years ago - 15 comments

#20 - out of memory

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

#19 - Fix command for model training

Pull Request - State: closed - Opened by yenchenlin over 6 years ago

#18 - Is possible to generate labels from facades images?

Issue - State: closed - Opened by happsky over 6 years ago - 1 comment

#17 - RuntimeError: size mismatch , when using my own 512x512 dataset

Issue - State: closed - Opened by lxj616 over 6 years ago - 2 comments

#16 - Can I use my own data for training?

Issue - State: closed - Opened by happsky over 6 years ago - 1 comment

#15 - About Encoder

Issue - State: closed - Opened by mathfinder over 6 years ago - 3 comments

#14 - Question about backward_G_alone

Issue - State: closed - Opened by eveningglow over 6 years ago - 1 comment

#13 - Should this be real_A_random instead of real_A_encoded?

Issue - State: closed - Opened by samrudhdhirangrej over 6 years ago - 3 comments

#12 - [Question] The bidirectional cycle-consistency losses advantage compared to pix2pix?

Issue - State: closed - Opened by lxj616 almost 7 years ago - 2 comments

#11 - A resize_ error

Issue - State: closed - Opened by JaheimLee almost 7 years ago - 1 comment

#10 - Fix unexpected keyword argument

Pull Request - State: closed - Opened by ufoym almost 7 years ago - 1 comment

GitHub / junyanz/BicycleGAN issues and pull requests