junyanz/pytorch-CycleGAN-and-pix2pix issues and pull requests

#1674 - RunTimeError : Size of Tensros must match except in dimension 1

Issue - State: open - Opened by Jiwno 8 days ago

#1673 - why I use BtoA in train.py and test.py, the ground truth is the B and img is A, then I get the fakeB?

Issue - State: open - Opened by master2021-bit 10 days ago

#1672 - Speed Up Training

Issue - State: open - Opened by wyd2 16 days ago - 1 comment

#1671 - there is no `latest_net_G.pth` in the directory where checkpoints are being saved

Issue - State: open - Opened by Rjaat 18 days ago - 2 comments

#1670 - light blue artifact on the output image

Issue - State: closed - Opened by XYZ-916 22 days ago

#1669 - input and output image size in training set

Issue - State: open - Opened by nanalaf about 2 months ago - 1 comment

#1668 - Rectangular image training failed，HELP！ HELP！

Issue - State: open - Opened by HolicAnn about 2 months ago

#1667 - What is the difference between preprocess scale_width_and_crop and resize_and_crop?

Issue - State: open - Opened by bclaw524 2 months ago

#1666 - Medical image

Issue - State: open - Opened by Xujingkk 2 months ago

#1665 - medical

Issue - State: open - Opened by Xujingkk 2 months ago

#1664 - inference of a single image (help)

Issue - State: open - Opened by themoshe12 3 months ago - 1 comment

#1663 - Use of pix2pix with multiple image types in training set

Issue - State: open - Opened by jasonpluce 4 months ago

#1662 - Query realted if the model can give RGBA image as output?

Issue - State: open - Opened by roshan2024nar 4 months ago - 2 comments

#1661 - Template-to-Image test output are blurry.

Issue - State: open - Opened by bclaw524 4 months ago

#1660 - how to train for two input image and one output image

Issue - State: closed - Opened by roshan2024nar 4 months ago - 2 comments

#1659 - Can‘t open the pix2pix train/test website to download

Issue - State: open - Opened by XXXXemberXXXX 4 months ago

#1658 - Help!

Issue - State: open - Opened by izhaolinger 4 months ago - 1 comment

#1657 - Image truncated error

Issue - State: open - Opened by johnnyv75 4 months ago

#1656 - The test results with pix2pix were poor

Issue - State: open - Opened by zengyanjie 4 months ago

#1655 - good

Issue - State: closed - Opened by wukaidi66 4 months ago

#1654 - value of lambda_identity not to change the color

Issue - State: open - Opened by rnqm 4 months ago

#1653 - Create P Lohith Reddy

Pull Request - State: open - Opened by lohithreddy0123 4 months ago

#1652 - Update cycle_gan_model.py

Pull Request - State: open - Opened by shikha-gist 5 months ago

#1651 - Update cycle_gan_model.py

Pull Request - State: open - Opened by shikha-gist 5 months ago

#1649 - Update CycleGan

Pull Request - State: open - Opened by Anshi2609 5 months ago

#1648 - Perform interpolation between Input and Output

Issue - State: open - Opened by obeidahmad 5 months ago

#1647 - Make docker image

Pull Request - State: closed - Opened by zhelinisivanesan 5 months ago

#1646 - Training Parameters or Architecture Settings Recommendations

Issue - State: closed - Opened by selimceylan 5 months ago

#1645 - Error in combine_A_and_B.py for TIFF-files

Issue - State: open - Opened by teakfi 5 months ago

#1644 - Input images are stretched when using test

Issue - State: open - Opened by KreesoAuriga 5 months ago - 1 comment

#1643 - I add the kaist_dataset.py file supporting (thermal -> RGB) or (RGB -> thermal) domain transformer.

Pull Request - State: closed - Opened by BeomsikCho 5 months ago

#1642 - Create my_dataset.py

Pull Request - State: open - Opened by andrea2399 5 months ago

#1641 - How to finish Scene Text Editing task using Pix2Pix

Issue - State: open - Opened by keepcoding123 6 months ago

#1640 - How can I train with multi-input like (A1 ,A2……)->B?

Issue - State: open - Opened by kuhhg 6 months ago - 1 comment

#1638 - Error testing my own dataset using pix2pix

Issue - State: open - Opened by Rentong2022 6 months ago

#1637 - Add ROI specific loss function to generator

Issue - State: open - Opened by Diogo-Valente2111 6 months ago

#1636 - introducing CycleGAN-Turbo and pix2pix-turbo

Issue - State: open - Opened by junyanz 6 months ago

#1635 - Hello,

Issue - State: closed - Opened by czh886 6 months ago - 1 comment

#1634 - 请问测试集应该下载保存到什么文件夹呢，每次都会报错'./checkpoints\\facades_label2photo_pretrained\\latest_net_G.pth'

Issue - State: open - Opened by xueyuan1126 6 months ago

#1633 - Is the Model Capable of Processing and Maintaining Consistent Output Sizes Across Varied Image Dimensions?

Issue - State: closed - Opened by arrrrr3186 6 months ago - 6 comments

#1632 - It gets stuck when running for more than 20 rounds, but the graphics card is still running. What is the reason for this?

Issue - State: open - Opened by LeeHaoRanRan 6 months ago

#1631 - CPU

Issue - State: open - Opened by AABBCCDKG 6 months ago - 1 comment

#1630 - 请问对于输入图像大小不一样时，该代码中是否对图像进行预处理了呢？

Issue - State: open - Opened by LauraABCD 6 months ago

#1629 - 生成图像的质量太低怎么办

Issue - State: open - Opened by LauraABCD 6 months ago

#1628 - 请问作者

Issue - State: open - Opened by LauraABCD 6 months ago - 1 comment

#1627 - 礼貌提问并虚心请教：线性衰减学习率政策

Issue - State: open - Opened by Fengzhennan 7 months ago

#1626 - SAR to Optical image generation

Issue - State: open - Opened by arrrrr3186 7 months ago - 1 comment

#1625 - Loading issue in trained model

Issue - State: closed - Opened by arrrrr3186 7 months ago - 1 comment

#1624 - CycleGAN model transferring the object rather than the style

Issue - State: open - Opened by nishanthnandakumar 7 months ago

#1623 - How to use test.py to test both directions for cyclegan?

Issue - State: closed - Opened by lamwilton 8 months ago - 1 comment

#1622 - oserror: errno 24 too many open files

Issue - State: open - Opened by w-yasuo 9 months ago - 1 comment

#1621 - Not able to load a pre trained weight that I have used for training of custom data.

Issue - State: open - Opened by AGRocky 9 months ago - 5 comments

#1620 - Exception in user code

Issue - State: closed - Opened by L-1027 9 months ago

#1619 - V aug 2023

Pull Request - State: closed - Opened by sebasmos 10 months ago

#1618 - merged to main

Pull Request - State: closed - Opened by sebasmos 10 months ago

#1617 - How to generate an image and its corresponding mask at the same time?

Issue - State: open - Opened by EudicL 10 months ago

#1610 - Is it normal to use 2 black and white images?

Issue - State: open - Opened by Keiser04 11 months ago - 7 comments

#1609 - Generate photorealistic images from human-drawn sketches using GANs - any model suggestions?

Issue - State: open - Opened by ehandorfer 11 months ago - 1 comment

#1608 - wandb & visdom question

Issue - State: open - Opened by xia-QQ 11 months ago - 1 comment

#1600 - poor test results - > recommendations for improvement

Issue - State: open - Opened by shxnying 12 months ago - 1 comment

#1597 - How to improve the low qulity image detail?

Issue - State: open - Opened by JunCS1 about 1 year ago - 1 comment

#1595 - Training loss does not decrease

Issue - State: open - Opened by nguyenpbui about 1 year ago - 3 comments

#1594 - pix2pix training doesn't go past "create web directory ./checkpoints\cartoonify_pix2pix\web..."

Issue - State: closed - Opened by TehreemFarooqi about 1 year ago - 2 comments

#1585 - Training pix2pix with different input size

Issue - State: open - Opened by MrMonk3y about 1 year ago - 3 comments

#1579 - HELP： Error(s) in loading state_dict for ResnetGenerator:

Issue - State: open - Opened by tjnkyqcy over 1 year ago - 5 comments

#1572 - Pix2Pix Input Inference Image Looks Different From the Original Inference Image

Issue - State: open - Opened by chokevin8 over 1 year ago - 2 comments

#1557 - save output as '.npy' file

Issue - State: closed - Opened by Ivvvvvvvvvvy over 1 year ago - 1 comment

#1528 - Cannot find reference 'create_model' in 'imported module models'

Issue - State: closed - Opened by mengxianglong123 over 1 year ago - 1 comment

#1514 - [multi-input for cycleGAN]

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

#1490 - CUDA version

Issue - State: open - Opened by Ziyan-swu almost 2 years ago - 2 comments

#1476 - AttributeError: 'Sequential' object has no attribute 'model'

Issue - State: closed - Opened by wang-TJ-20 about 2 years ago - 2 comments

#1450 - AttributeError: module 'torchvision.transforms' has no attribute 'InterpolationMode'

Issue - State: closed - Opened by DaoLinZhou about 2 years ago - 8 comments

#1446 - using pre-trained day2night for own dataset

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

#1440 - Poor pix2pixHD and pix2pix test results

Issue - State: open - Opened by songyn95 over 2 years ago - 4 comments

#1419 - Training Freezes

Issue - State: open - Opened by aakash-chakraborty1995 over 2 years ago - 6 comments

#1412 - UserWarning: Detected call of `lr_scheduler.step()` before `optimizer.step()`.

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

#1395 - continue training

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

#1369 - how to set multiple GPU running the train.py ？

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

#1301 - Images are getting predicted with inverted color

Issue - State: open - Opened by souryasengupta about 3 years ago - 3 comments

#1248 - Issue with combine_A_and_B

Issue - State: open - Opened by sinievanderben over 3 years ago - 8 comments

#1214 - RuntimeError: Error(s) in loading state_dict for UnetGenerator:

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

#1095 - Continue/ resume training: update learning rate

Issue - State: closed - Opened by Bowen-Lu about 4 years ago - 7 comments

#1059 - Failing to reconstruct high value pixels with pix2pix

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

#1057 - Convert to fake image badly but reconstruct well

Issue - State: closed - Opened by poppopting over 4 years ago - 11 comments

#977 - Unet generator for large images

Issue - State: open - Opened by myfrannie over 4 years ago - 6 comments

#902 - Should we use eval for discriminator in the training of generator if dropout is used?

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

#827 - Generate infrared images from visual images

Issue - State: closed - Opened by MuhammadAsadJaved almost 5 years ago - 3 comments

#766 - How to alternate between training and validating?

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

#764 - Training Time

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

#687 - loading model error -- AttributeError: 'UnetSkipConnectionBlock'

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

#668 - Confusion with CycleGAN testing regarding domains

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

#624 - Feature request: SSIM loss

Issue - State: closed - Opened by John1231983 over 5 years ago - 2 comments

#548 - Training is good but testing has a problem.

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

#518 - Generated Images have noise in them

Issue - State: closed - Opened by shreyasbapat over 5 years ago - 12 comments

#478 - help with the colorization dataset mode

Issue - State: open - Opened by ajaykmr2292 over 5 years ago - 11 comments

#439 - Wasserstein distance?

Issue - State: closed - Opened by Arkkienkeli almost 6 years ago - 27 comments

GitHub / junyanz/pytorch-CycleGAN-and-pix2pix issues and pull requests