代码：
accelerate launch examples/boogu_image/model_training/train.py
--dataset_base_path "/workspace/lk/model/llm/multimodal/image_lora"
--dataset_metadata_path "/workspace/lk/model/llm/multimodal/image_lora/metadata_boogu.csv"
--max_pixels 1048576
--dataset_repeat 5
--model_id_with_origin_paths "dit:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/transformer/.safetensors,mllm:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/mllm/.safetensors,vae:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/vae/*.safetensors"
--processor_path "/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit"
--learning_rate 1e-4
--num_epochs 5
--remove_prefix_in_ckpt "pipe.dit."
--output_path "/workspace/lk/output/vlm/Boogu-Image-0.1-Edit_lora"
--lora_base_model "dit"
--lora_target_modules "to_q,to_k,to_v,to_out.0,img_to_q,img_to_k,img_to_v,img_out,instruct_to_q,instruct_to_k,instruct_to_v,instruct_out"
--lora_rank 32
--use_gradient_checkpointing
--dataset_num_workers 8
--find_unused_parameters
--data_file_keys "image,edit_image,prompt"
--extra_inputs "edit_image"
报错：
Traceback (most recent call last):
File "/workspace/lk/model/server/DiffSynth-Studio/examples/boogu_image/model_training/train.py", line 106, in
model = BooguImageTrainingModule(
File "/workspace/lk/model/server/DiffSynth-Studio/examples/boogu_image/model_training/train.py", line 29, in init
self.pipe = BooguImagePipeline.from_pretrained(torch_dtype=torch.bfloat16, device=device, model_configs=model_configs, processor_config=processor_config)
File "/workspace/lk/model/server/DiffSynth-Studio/diffsynth/pipelines/boogu_image.py", line 59, in from_pretrained
model_pool = pipe.download_and_load_models(model_configs, vram_limit)
File "/workspace/lk/model/server/DiffSynth-Studio/diffsynth/diffusion/base_pipeline.py", line 300, in download_and_load_models
model_config.download_if_necessary()
File "/workspace/lk/model/server/DiffSynth-Studio/diffsynth/core/loader/config.py", line 100, in download_if_necessary
self.download()
File "/workspace/lk/model/server/DiffSynth-Studio/diffsynth/core/loader/config.py", line 66, in download
snapshot_download(
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/modelscope/hub/snapshot_download.py", line 108, in snapshot_download
return _snapshot_download(
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/modelscope/hub/snapshot_download.py", line 254, in _snapshot_download
endpoint = _api.get_endpoint_for_read(
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/modelscope/hub/api.py", line 332, in get_endpoint_for_read
if not self.repo_exists(
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/modelscope/hub/api.py", line 379, in repo_exists
raise Exception('Invalid repo_id: %s, must be of format namespace/name' % repo_type)
Exception: Invalid repo_id: model, must be of format namespace/name
Traceback (most recent call last):
File "/root/miniforge3/envs/diffSynth_studio/bin/accelerate", line 8, in
sys.exit(main())
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/accelerate/commands/accelerate_cli.py", line 50, in main
args.func(args)
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/accelerate/commands/launch.py", line 1235, in launch_command
simple_launcher(args)
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/accelerate/commands/launch.py", line 823, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/root/miniforge3/envs/diffSynth_studio/bin/python', 'examples/boogu_image/model_training/train.py', '--dataset_base_path', '/workspace/lk/model/llm/multimodal/image_lora', '--dataset_metadata_path', '/workspace/lk/model/llm/multimodal/image_lora/metadata_boogu.csv', '--max_pixels', '1048576', '--dataset_repeat', '5', '--model_id_with_origin_paths', 'dit:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/transformer/.safetensors,mllm:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/mllm/.safetensors,vae:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/vae/*.safetensors', '--processor_path', '/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit', '--learning_rate', '1e-4', '--num_epochs', '5', '--remove_prefix_in_ckpt', 'pipe.dit.', '--output_path', '/workspace/lk/output/vlm/Boogu-Image-0.1-Edit_lora', '--lora_base_model', 'dit', '--lora_target_modules', 'to_q,to_k,to_v,to_out.0,img_to_q,img_to_k,img_to_v,img_out,instruct_to_q,instruct_to_k,instruct_to_v,instruct_out', '--lora_rank', '32', '--use_gradient_checkpointing', '--dataset_num_workers', '8', '--find_unused_parameters', '--data_file_keys', 'image,edit_image,prompt', '--extra_inputs', 'edit_image']' returned non-zero exit status 1.
代码：
accelerate launch examples/boogu_image/model_training/train.py
--dataset_base_path "/workspace/lk/model/llm/multimodal/image_lora"
--dataset_metadata_path "/workspace/lk/model/llm/multimodal/image_lora/metadata_boogu.csv"
--max_pixels 1048576
--dataset_repeat 5
--model_id_with_origin_paths "dit:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/transformer/.safetensors,mllm:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/mllm/.safetensors,vae:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/vae/*.safetensors"
--processor_path "/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit"
--learning_rate 1e-4
--num_epochs 5
--remove_prefix_in_ckpt "pipe.dit."
--output_path "/workspace/lk/output/vlm/Boogu-Image-0.1-Edit_lora"
--lora_base_model "dit"
--lora_target_modules "to_q,to_k,to_v,to_out.0,img_to_q,img_to_k,img_to_v,img_out,instruct_to_q,instruct_to_k,instruct_to_v,instruct_out"
--lora_rank 32
--use_gradient_checkpointing
--dataset_num_workers 8
--find_unused_parameters
--data_file_keys "image,edit_image,prompt"
--extra_inputs "edit_image"
报错：
Traceback (most recent call last):
File "/workspace/lk/model/server/DiffSynth-Studio/examples/boogu_image/model_training/train.py", line 106, in
model = BooguImageTrainingModule(
File "/workspace/lk/model/server/DiffSynth-Studio/examples/boogu_image/model_training/train.py", line 29, in init
self.pipe = BooguImagePipeline.from_pretrained(torch_dtype=torch.bfloat16, device=device, model_configs=model_configs, processor_config=processor_config)
File "/workspace/lk/model/server/DiffSynth-Studio/diffsynth/pipelines/boogu_image.py", line 59, in from_pretrained
model_pool = pipe.download_and_load_models(model_configs, vram_limit)
File "/workspace/lk/model/server/DiffSynth-Studio/diffsynth/diffusion/base_pipeline.py", line 300, in download_and_load_models
model_config.download_if_necessary()
File "/workspace/lk/model/server/DiffSynth-Studio/diffsynth/core/loader/config.py", line 100, in download_if_necessary
self.download()
File "/workspace/lk/model/server/DiffSynth-Studio/diffsynth/core/loader/config.py", line 66, in download
snapshot_download(
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/modelscope/hub/snapshot_download.py", line 108, in snapshot_download
return _snapshot_download(
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/modelscope/hub/snapshot_download.py", line 254, in _snapshot_download
endpoint = _api.get_endpoint_for_read(
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/modelscope/hub/api.py", line 332, in get_endpoint_for_read
if not self.repo_exists(
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/modelscope/hub/api.py", line 379, in repo_exists
raise Exception('Invalid repo_id: %s, must be of format namespace/name' % repo_type)
Exception: Invalid repo_id: model, must be of format namespace/name
Traceback (most recent call last):
File "/root/miniforge3/envs/diffSynth_studio/bin/accelerate", line 8, in
sys.exit(main())
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/accelerate/commands/accelerate_cli.py", line 50, in main
args.func(args)
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/accelerate/commands/launch.py", line 1235, in launch_command
simple_launcher(args)
File "/root/miniforge3/envs/diffSynth_studio/lib/python3.10/site-packages/accelerate/commands/launch.py", line 823, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/root/miniforge3/envs/diffSynth_studio/bin/python', 'examples/boogu_image/model_training/train.py', '--dataset_base_path', '/workspace/lk/model/llm/multimodal/image_lora', '--dataset_metadata_path', '/workspace/lk/model/llm/multimodal/image_lora/metadata_boogu.csv', '--max_pixels', '1048576', '--dataset_repeat', '5', '--model_id_with_origin_paths', 'dit:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/transformer/.safetensors,mllm:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/mllm/.safetensors,vae:/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit/vae/*.safetensors', '--processor_path', '/workspace/lk/model/llm/vlm/Boogu-Image-01-Edit', '--learning_rate', '1e-4', '--num_epochs', '5', '--remove_prefix_in_ckpt', 'pipe.dit.', '--output_path', '/workspace/lk/output/vlm/Boogu-Image-0.1-Edit_lora', '--lora_base_model', 'dit', '--lora_target_modules', 'to_q,to_k,to_v,to_out.0,img_to_q,img_to_k,img_to_v,img_out,instruct_to_q,instruct_to_k,instruct_to_v,instruct_out', '--lora_rank', '32', '--use_gradient_checkpointing', '--dataset_num_workers', '8', '--find_unused_parameters', '--data_file_keys', 'image,edit_image,prompt', '--extra_inputs', 'edit_image']' returned non-zero exit status 1.