wpl427

Results 2 comments of wpl427

Or how to reduce video memory usage by modifying the configuration?

before: prompt_model,_,_ = open_clip.create_model_and_transforms('ViT-bigG-14', 'laion2b_s39b_b160k') after: prompt_model,_,_ = open_clip.create_model_and_transforms('ViT-bigG-14', 'laion2b_s39b_b160k',precision='fp16') error: (faceswap) [root@prod-emr-gpu01 StyleDrop-PyTorch]# accelerate launch --num_processes 8 --mixed_precision fp16 train_t2i_custom_v2.py --config=configs/custom.py 2023-08-21 13:16:44.463755: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is...