DCR

DCR copied to clipboard

python diff_inference.py get OutOfMemoryError

2

I first generated the pictures using the diff_inference.py `python diff_inference.py -nb 4000 --dataset laion --capstyle instancelevel_blip --rand_augs rand_numb_add` while I met File "/home/anaconda3/envs/diffrep/lib/python3.9/site-packages/diffusers/models/cross_attention.py", line 314, in __call__ attention_probs = attn.get_attention_scores(query,...

whybfq

fine tune model error

2

accelerate launch diff_train.py \ --pretrained_model_name_or_path stabilityai/stable-diffusion-2-1 \ --instance_data_dir train/images_large \ --resolution=256 --gradient_accumulation_steps=1 --center_crop --random_flip \ --learning_rate=5e-6 --lr_scheduler constant_with_warmup \ --lr_warmup_steps=5000 --max_train_steps=100000 \ --train_batch_size=16 --save_steps=10000 --modelsavesteps 20000 --duplication nodup \ --output_dir=output...

whybfq

missing pkgs

1

Hi The following packages need to be installed too scikit-image seaborn timm LovelyPlots matplotlib natsort scikit-learn triton Also CLIP is installed via pip in env.yaml, but https://github.com/openai/CLIP/issues/329#issuecomment-1439676212 that should be...

rahimentezari

Thank you very much for your work. Could the 1200 duplicated images mentioned in the paper be released?

yueanga