VITA-Group/TransGAN issues and pull requests

#89 - Shape mismatch: inputs 32 != attn_output 1024

Issue - State: open - Opened by code-Kasim about 1 month ago

#88 - Request to Update "Representative Visual Results" Section

Issue - State: open - Opened by wuyu1212 about 1 year ago

#87 - load the .pkl file error

Issue - State: open - Opened by JiangYan0409 about 1 year ago

#86 - About Model

Issue - State: open - Opened by journeyMe1 over 1 year ago

#85 - question about argumentation

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

#84 - Some question about loss

Issue - State: closed - Opened by JUSTM0VE0N over 1 year ago - 1 comment

#83 - about test

Issue - State: open - Opened by maoshen5 over 2 years ago - 10 comments

#82 - About the Patch Splitting Image.

Issue - State: closed - Opened by GajuuzZ over 2 years ago - 3 comments

#81 - Could you give us the CelebA pretrained model, please?

Issue - State: open - Opened by zzr525 over 2 years ago - 1 comment

#80 - Representative Visual Results

Issue - State: open - Opened by zzr525 over 2 years ago

#79 - train

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

#78 - test error

Issue - State: open - Opened by maoshen5 over 2 years ago - 1 comment

#77 - Fid stat

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

#76 - about Attention scale

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

#75 - One error

Issue - State: open - Opened by wudiduojimone over 2 years ago - 7 comments

#74 - 请问有生成器是transgan，判别器是autogan的超参数吗

Issue - State: closed - Opened by maoshen5 over 2 years ago - 2 comments

#73 - A question about training

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

#72 - ValueError: Invalid backend: ''nccl''

Issue - State: closed - Opened by Jamie-Cheung over 2 years ago - 2 comments

#71 - Reporting a minor bug

Issue - State: open - Opened by hubert0527 almost 3 years ago - 1 comment

#51 - g_loss

Issue - State: closed - Opened by Stephenlove almost 3 years ago - 1 comment

#50 - Using Conv?

Issue - State: closed - Opened by jongwook-Han about 3 years ago - 2 comments

#49 - FID score

Issue - State: closed - Opened by guoliangq about 3 years ago - 8 comments

#48 - run Transgan with single gpu

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

#47 - RuntimeError: The size of tensor a (5) must match the size of tensor b (4097) at non-singleton dimension 1

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

#46 - Question about ViT_custom_local544444_256_rp.py

Issue - State: closed - Opened by RSMung about 3 years ago - 2 comments

#45 - Question about GAN training.

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

#44 - model mismatch

Issue - State: closed - Opened by yallien about 3 years ago - 4 comments

#43 - Custom dataset HOW-TO and is this capable of conditional GAN ?

Issue - State: open - Opened by thusinh1969 about 3 years ago - 6 comments

#42 - Could you please provide the celeba-hq-256 checkpoint?

Issue - State: open - Opened by yzcv about 3 years ago - 2 comments

#41 - The trick of exponential moving average weights for the generator

Issue - State: closed - Opened by bad-meets-joke over 3 years ago - 1 comment

#40 - The result FID on CIFAR10 could not be reimplemented.

Issue - State: closed - Opened by bad-meets-joke over 3 years ago - 3 comments

#39 - Question about Output of dimension

Issue - State: closed - Opened by ygjwd12345 over 3 years ago - 3 comments

#38 - Training for CIFAR-10

Issue - State: closed - Opened by Andrew-X-Wang over 3 years ago - 2 comments

#37 - celeba

Issue - State: closed - Opened by chalure over 3 years ago - 1 comment

#36 - Generated Images Have Some Blocking Artifact

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

#35 - Single GPU running model

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

#34 - The model code

Issue - State: closed - Opened by LixDemon over 3 years ago - 1 comment

#33 - 你好，请问我应该如何复现论文效果，例如cifar10上的效果，如何加载预训练模型？是使用fid stat中的文件么

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

#32 - About the patchsize in the Generator

Issue - State: closed - Opened by destructive-observer over 3 years ago - 16 comments

#31 - Error occurs when training on Cifar-10

Issue - State: closed - Opened by wind754203900 over 3 years ago - 3 comments

#30 - Training on single GPU

Issue - State: closed - Opened by thearkaprava over 3 years ago - 10 comments

#29 - UnboundLocalError: local variable 'gen_net' referenced before assignment

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

#28 - Model Size

Issue - State: open - Opened by Hsintien-Ng over 3 years ago

#27 - which is the defualt gen_model

Issue - State: closed - Opened by AragornKris over 3 years ago - 1 comment

#26 - celeba image size

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

#25 - Linear Unflatten layer

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

#24 - models_search

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

#23 - fid

Issue - State: closed - Opened by chalure over 3 years ago - 1 comment

#22 - fid_stats_cifar10_train.npz

Issue - State: closed - Opened by zhangwei025 over 3 years ago - 1 comment

#21 - Where is implementation of MT-CT?

Issue - State: closed - Opened by subminu over 3 years ago - 1 comment

#20 - Is the function `get_attn_mask` the same one used in the paper on arxiv?

Issue - State: closed - Opened by azfarkhoja305 over 3 years ago - 2 comments

#19 - Could you please provide a training script? very appreciate！

Issue - State: closed - Opened by imkelt over 3 years ago - 1 comment

#18 - loss

Issue - State: closed - Opened by Fengww2014 over 3 years ago - 4 comments

#17 - How finetune the code on my own dataset?

Issue - State: closed - Opened by nnajeh over 3 years ago - 1 comment

#16 - training code

Issue - State: closed - Opened by GaranWu over 3 years ago - 2 comments

#15 - Please advise on how to use text versus images as data

Issue - State: closed - Opened by dumbjarvis over 3 years ago - 1 comment

#14 - Self-Attention

Issue - State: closed - Opened by ilyak93 over 3 years ago - 1 comment

#13 - Can this model generate 128 or 256 resolution images?

Issue - State: closed - Opened by DaddyJin over 3 years ago - 1 comment

#12 - train

Issue - State: closed - Opened by GooDrYu over 3 years ago - 2 comments

#11 - which one is the cifar10 checkpoint

Issue - State: closed - Opened by lingcong-k over 3 years ago - 3 comments

#10 - Question

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

#9 - About the role of function “get_attn_mask”.

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

#8 - Can generate 1024x1024 images?

Issue - State: closed - Opened by wodsoe over 3 years ago - 2 comments

#7 - Training fails

Issue - State: closed - Opened by anj1 almost 4 years ago - 8 comments

#6 - longformer model not present

Issue - State: closed - Opened by anj1 almost 4 years ago

#5 - Image sizes don't match

Issue - State: closed - Opened by anj1 almost 4 years ago - 2 comments

#4 - Checkpoint doesn't match

Issue - State: closed - Opened by anj1 almost 4 years ago - 3 comments

#3 - Training script

Issue - State: closed - Opened by maximzubkov almost 4 years ago - 7 comments

#2 - closed

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

#1 - Evolution of the generations during training

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

GitHub / VITA-Group/TransGAN issues and pull requests