CompVis/taming-transformers issues and pull requests

#258 - VQGAN Training details

Issue - State: open - Opened by 649459021 17 days ago

#257 - Training transformer on custom data configuration file is missing

Issue - State: open - Opened by RAHULPRRAHUL 18 days ago

#256 - LearningRateMonitor

Issue - State: open - Opened by Echo-jyt about 1 month ago

#255 - How do I sample after training a model on a custom dataset?

Issue - State: open - Opened by Echo-jyt about 2 months ago

#254 - Update main.py

Pull Request - State: open - Opened by MaxxP0 2 months ago

#253 - A question regarding the embed_dim parameter in the config file.

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

#252 - Cederic

Pull Request - State: closed - Opened by cederican 3 months ago

#251 - : Parameter indices which did not receive grad for rank 0: 338 339 340 341 342 343 344 345 346 347 348 349 350

Issue - State: open - Opened by ashwani-ver 4 months ago - 3 comments

#250 - generate white images during training

Issue - State: open - Opened by NikLi66 4 months ago - 1 comment

#249 - Confusion about the dimension of latent vectors

Issue - State: open - Opened by LiJiahao-Alex 4 months ago

#248 - ncc

Issue - State: closed - Opened by Faded1022 4 months ago - 1 comment

#247 - ModuleNotFoundError: No module named 'cutlit'

Issue - State: open - Opened by Polaris0421 5 months ago

#246 - Create init.py inside taming directory

Pull Request - State: closed - Opened by tarun018 5 months ago

#245 - Is there a plan to upgrade this repo to Pytorch 2.0

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

#244 - How to use the sliding attention window mechanism？

Issue - State: open - Opened by RichardXue123 6 months ago - 3 comments

#243 - Fix variable naming errors in EMAVectorQuantizer

Pull Request - State: open - Opened by JJJYmmm 6 months ago

#242 - How to distinguish sos token(default = 0) and quantified image token zero ?

Issue - State: open - Opened by JJJYmmm 6 months ago - 2 comments

#241 - How to train your own Transformer

Issue - State: open - Opened by zxb-0 7 months ago

#240 - 'VQModel' object has no attribute 'transformer'

Issue - State: open - Opened by Gannn12138 7 months ago

#239 - Dockerfile/environment.yaml for CUDA Version 12.3?

Issue - State: closed - Opened by kajc10 8 months ago - 5 comments

#238 - README Error -- yaml path

Issue - State: open - Opened by catsled 8 months ago

#237 - When I train, the code is getting slower and slower, is it normal?

Issue - State: open - Opened by zxb-0 8 months ago - 1 comment

#236 - Where is the model trained on my custom data?

Issue - State: open - Opened by zxb-0 8 months ago - 1 comment

#235 - gpu training hangs

Issue - State: open - Opened by shyakocat 8 months ago - 2 comments

#234 - Cannot use gpu to fine-tune on custom-data

Issue - State: open - Opened by YkiWu 9 months ago - 2 comments

#233 - If my condition is class label, how to set config file

Issue - State: open - Opened by Ontheroad123 9 months ago

#232 - How to train the transformer after training the vqgan model？

Issue - State: closed - Opened by RichardXue123 9 months ago - 9 comments

#231 - typo fix README.md master

Pull Request - State: open - Opened by Anner-deJong 11 months ago

#230 - Add init.py to make pip installable modules

Pull Request - State: open - Opened by cadolphs 11 months ago

#229 - How to do image inpainting with vqgan?

Issue - State: open - Opened by cannonli7 11 months ago

#228 - Simply concatenating the source indices and condition indices even when they are from different codebooks?

Issue - State: open - Opened by abcqmars 12 months ago - 2 comments

#227 - NCCL error

Issue - State: closed - Opened by alex-pv01 12 months ago - 2 comments

#226 - Fix/taming download path

Pull Request - State: open - Opened by shibuiwilliam about 1 year ago

#225 - Update main.py

Pull Request - State: closed - Opened by sandrodevdariani about 1 year ago

#224 - Where to get the config file of VQGAN OpenImages (f=8), 256?

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

#223 - ModuleNotFoundError: No module named 'torch._six'

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

#222 - Stable diffusion VAE fine tuning (backport AutoencoderKL and its config.yaml to taming-transformers)

Pull Request - State: open - Opened by rbbb about 1 year ago - 3 comments

#221 - small question about the vq-vae paper

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

#220 - Invalid version: '0.10.1,<0.11'

Issue - State: closed - Opened by EinfachYan about 1 year ago - 7 comments

#219 - Vector Quantization loss rises in training

Issue - State: closed - Opened by OrangeSodahub about 1 year ago - 3 comments

#218 - [Code Error]

Issue - State: closed - Opened by haozhi1817 about 1 year ago

#217 - Create the last stand

Pull Request - State: closed - Opened by TheCamelWhisperer420 about 1 year ago

#216 - eleba

Issue - State: closed - Opened by tsa18 about 1 year ago

#215 - Openimages version.

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

#214 - Implementation question - Resize then crop - Model trained on low resolution image?

Issue - State: open - Opened by HieuPhan33 about 1 year ago - 2 comments

#213 - AttributeError: module 'signal' has no attribute 'SIGUSR1'

Issue - State: open - Opened by nhw649 about 1 year ago - 1 comment

#212 - about loss of training stage2 transformers

Issue - State: open - Opened by CrossLee1 over 1 year ago - 8 comments

#211 - Drop nonexisting module torch._six

Pull Request - State: closed - Opened by olaviinha over 1 year ago - 1 comment

#210 - wrong enviroment file

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

#209 - VQGAN models license

Issue - State: open - Opened by Aspie96 over 1 year ago

#208 - How to calculate the negative log-likelihood from the first stage VQGAN model?

Issue - State: open - Opened by khanhvu207 over 1 year ago

#207 - Hi! pretrained VQGAN face model!

Issue - State: open - Opened by scj0709 over 1 year ago

#206 - results of vqgan

Issue - State: open - Opened by Winnie202 over 1 year ago - 2 comments

#205 - No module named 'torch._six'

Issue - State: open - Opened by Gianluca124 over 1 year ago - 6 comments

#204 - How to train vqgan on coco dataset?

Issue - State: open - Opened by kevinchow1993 over 1 year ago - 1 comment

#203 - i want to see taming.models.vqgan.VQModel train log,

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

#202 - torch._six is depreciated, remove import

Pull Request - State: open - Opened by miamannionx over 1 year ago - 3 comments

#201 - Have reduced typo README

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

#200 - Remove reference to deprecated feature

Pull Request - State: open - Opened by mlo0352 over 1 year ago

#199 - pip._internal.vcs.versioncontrol.RemoteNotFoundError

Issue - State: open - Opened by tianjun318 over 1 year ago - 2 comments

#198 - installing into conda virtualenv (not working?)

Issue - State: open - Opened by agi-codes over 1 year ago

#197 - FileNotFoundError: [Errno 2] No such file or directory: '/path/to/coco_epoch117.ckpt'

Issue - State: open - Opened by justinday123 over 1 year ago

#196 - How to get the url of checkpoint and config?

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

#195 - 'Labelator' object has no attribute 'decode'

Issue - State: open - Opened by halfbloodprincecode almost 2 years ago

#194 - pytorch still raised "out of memory" but my PYTORCH_CUDA_ALLOC_CONF = "max_split_size_mb:128"

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

#193 - Pretrained stage 2 VQGAN?

Issue - State: closed - Opened by manuelknott almost 2 years ago

#192 - CVE-2007-4559 Patch

Pull Request - State: open - Opened by TrellixVulnTeam almost 2 years ago

#191 - how does the quantization book avoid trivial solution?

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

#190 - Use compatible import

Pull Request - State: closed - Opened by carmocca almost 2 years ago - 1 comment

#189 - pytorch light issue

Issue - State: closed - Opened by willyk26 almost 2 years ago - 3 comments

#188 - Update rank_zero_only after pytorch-lightening breaking change

Pull Request - State: closed - Opened by apolinario almost 2 years ago

#187 - Configs for some models

Issue - State: open - Opened by netw0rkf10w almost 2 years ago - 1 comment

#186 - code for classifier-guided rejection sampling

Issue - State: open - Opened by xinmiaolin almost 2 years ago

#185 - Did lr(learning rate) scheduler was used?

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

#184 - Can't train used to multi gpu

Issue - State: open - Opened by ryhhtn almost 2 years ago - 3 comments

#183 - The use of the parameter num_res_blocks is different between the encoder and the decoder

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

#182 - Missing init.py and relative import

Pull Request - State: open - Opened by KernelA almost 2 years ago

#181 - Conditional image generation

Issue - State: open - Opened by ChaoyiZh almost 2 years ago

#180 - Syntax Error

Issue - State: open - Opened by Jack1e940625 almost 2 years ago

#179 - Saving multiple checkpoints

Issue - State: closed - Opened by GiannisPikoulis almost 2 years ago - 1 comment

#178 - How do I set parameter f to a specified number?

Issue - State: open - Opened by Mengzibin almost 2 years ago

#177 - upload CelebA-HQ-256

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

#176 - ModuleNotFoundError: no module named "taming"

Issue - State: open - Opened by Nughu about 2 years ago - 9 comments

#175 - [Request Information] Where to find the trained model?

Issue - State: open - Opened by aamir09 about 2 years ago

#174 - Where to find configs for VQGAN OpenImages (f=8), 16384

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

#173 - Support pip install

Pull Request - State: open - Opened by illeatmyhat about 2 years ago

#172 - batch size in conf file is set for one GPU or all GPU?

Issue - State: open - Opened by fido20160817 about 2 years ago

#171 - how to train a images whose h and w different?

Issue - State: open - Opened by fido20160817 about 2 years ago

#170 - set resolution and size to 1024 if I want to generate high resolution images?

Issue - State: open - Opened by fido20160817 about 2 years ago - 1 comment

#169 - first stage model of celeba?

Issue - State: open - Opened by fido20160817 about 2 years ago - 1 comment

#168 - Allow folders for the train and validation paths.

Pull Request - State: open - Opened by ZeroCool940711 about 2 years ago

#167 - Only the images of the first five epochs were logged.

Issue - State: open - Opened by adsf0427 about 2 years ago

#166 - How to set condition stage modelto be the same with the first stage in config file?

Issue - State: closed - Opened by adsf0427 about 2 years ago

#165 - About diffusion modules

Issue - State: closed - Opened by Asthestarsfalll about 2 years ago

GitHub / CompVis/taming-transformers issues and pull requests