lucidrains/DALLE-pytorch
Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch
PythonMIT
Issues
- 4
Pretrained dalle.pt model file
#324 opened by tianye2856 - 1
Using pre-trained text-to-text language model in place of transformer part of the DALLE model.
#360 opened by Vbansal21 - 1
- 4
- 11
(colab notebook) Train DALLE-pytorch on C@H
#291 opened by afiaka87 - 3
Invalid PTX code
#378 opened by hamdjalil - 5
- 1
DALLE generating ugly image
#384 opened by SeungyounShin - 3
Hierarchical (Hourglass) Transformer
#382 opened by nsankar - 1
A link to dalle.pt that works for generate.py
#373 opened by IssamLaradji - 0
- 1
Help wanted: sequence length error
#374 opened by TatianaSnauwaert - 7
RuntimeError: Expected object of scalar type Half but got scalar type Float for argument #2 'mat2' in call to _th_bmm_out
#333 opened by Gitsamshi - 4
Running in Colab: AttributeError: module 'keras.backend' has no attribute 'is_tensor'
#356 opened by guillermoAlv - 1
--taming should ignore vae_params
#370 opened by TomorrowToday - 2
Bug in CLIP: image_fmap_size is None
#372 opened by jbdel - 3
Got error '' Input tensor shape: torch.Size([1024]). Additional info: {'b': 4}. Expected 2 dimensions, got 1'' everything used to work fine now it keeps giving me this error, it only happens on taming.
#326 opened by johngore123 - 16
New error using the new update.
#359 opened by jaoeded - 7
- 0
Generations taken from Train Set (CUB200 dataset)
#355 opened by SerezD - 8
Train with taming's VQGAN
#328 opened by WocMaker - 6
Missing keys while testing pre-trained models
#352 opened by Radeeswar - 2
How to cite your work.
#344 opened by WocMaker - 0
when generate imagas, error : __init__() got an unexpected keyword argument 'device'
#343 opened by zxy2020 - 4
- 0
- 7
- 1
Out of memory when using train_vae.py
#331 opened by soloist97 - 2
Why in training, the length of transformer token may exceeds the total text + image length?
#337 opened by cientgu - 0
20 Epochs on COCO - (Larger Transformer)
#336 opened by afiaka87 - 1
DO YOU KNOW YOUR DATASETS?
#334 opened by asigalov61 - 5
- 12
- 2
Bump the docker version for sparse attention?
#298 opened by afiaka87 - 6
Mannequins/Blogpost Trained with 16K VQGAN
#319 opened by afiaka87 - 0
Add top-p filtering
#318 opened by bmtm - 1
- 6
CogView Released! (But it requires a chinese phone number to download the weights)
#306 opened by afiaka87 - 4
Support 16k Token Imagenet VQGAN
#290 opened by afiaka87 - 3
sparse attn setting: ('full', 'sparse') required?
#294 opened by richcmwang - 9
Inference with DeepSpeed
#295 opened by afiaka87 - 16
CI / Tests for DALLE-pytorch
#301 opened by robvanvolt - 1
train dVAE lr scheduler step in wrong place?
#278 opened by yangyixiao - 1
Image mask should be an argument to generate.py
#300 opened by afiaka87 - 1
Big Bird Sparse Attention/Sparse Softmax
#292 opened by afiaka87 - 0
- 3
Training Broken Without Cached VQGAN
#281 opened by afiaka87 - 0
Channels
#282 opened by ethancohen123 - 4
deepspeed only uses a single GPU
#279 opened by richcmwang - 0
`--zero_stage <0|1|2|3>` flag for training
#276 opened by afiaka87