Open weimingma opened 3 weeks ago
To train an IP-Adapter_plus model, are the following commands correct especially the 2 model paths pretrained_model_name_or_path and image_encoder_path:
accelerate launch --num_processes 2 --multi_gpu --mixed_precision "fp16" \ tutorial_train_plus.py \ --pretrained_model_name_or_path="stable-diffusion-v1-5/" \ --image_encoder_path="models/image_encoder/" \ --data_json_file="assets/prompt_image.json" \ --data_root_path="assets/train/" \ --mixed_precision="fp16" \ --resolution=512 \ --train_batch_size=2 \ --dataloader_num_workers=4 \ --learning_rate=1e-04 \ --weight_decay=0.01 \ --output_dir="out_model/" \ --save_steps=3
不对，把模型下载下来用本地路径
wx13966113394
commented
2 weeks ago
To train an IP-Adapter_plus model, are the following commands correct especially the 2 model paths pretrained_model_name_or_path and image_encoder_path:
accelerate launch --num_processes 2 --multi_gpu --mixed_precision "fp16" \ tutorial_train_plus.py \ --pretrained_model_name_or_path="stable-diffusion-v1-5/" \ --image_encoder_path="models/image_encoder/" \ --data_json_file="assets/prompt_image.json" \ --data_root_path="assets/train/" \ --mixed_precision="fp16" \ --resolution=512 \ --train_batch_size=2 \ --dataloader_num_workers=4 \ --learning_rate=1e-04 \ --weight_decay=0.01 \ --output_dir="out_model/" \ --save_steps=3