Upload metadata/low_high_cost_8gpu_latent_text_continue_si/real_train_status.json with huggingface_hub

metadata/low_high_cost_8gpu_latent_text_continue_si/real_train_status.json

@@ -0,0 +1,86 @@
+{
+  "backend": "ditto",
+  "launcher": "accelerate",
+  "accelerate_num_processes": 8,
+  "accelerate_config_file": "/mnt/si002961ale4/default/lgy/shiying/low-high-new/scripts/training/edit-model/accelerate_config_8gpu_bf16.yaml",
+  "cwd": "/mnt/si002961ale4/default/lgy/shiying/low-high-new/Ditto",
+  "command": [
+    "/opt/conda/bin/python3",
+    "-m",
+    "accelerate.commands.launch",
+    "--config_file",
+    "/mnt/si002961ale4/default/lgy/shiying/low-high-new/scripts/training/edit-model/accelerate_config_8gpu_bf16.yaml",
+    "--num_processes",
+    "8",
+    "/mnt/si002961ale4/default/lgy/shiying/low-high-new/Ditto/examples/wanvideo/model_training/train.py",
+    "--dataset_base_path",
+    "/mnt/si002961ale4/default/lgy/shiying/low-high-new",
+    "--dataset_metadata_path",
+    "/mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/low_high_cost_8gpu_latent_text_continue_si/real_train/metadata.train.csv",
+    "--tokenizer_path",
+    "/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/google/umt5-xxl",
+    "--data_file_keys",
+    "video,vace_video",
+    "--height",
+    "480",
+    "--width",
+    "832",
+    "--num_frames",
+    "17",
+    "--dataset_repeat",
+    "1",
+    "--model_paths",
+    "[[\"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00001-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00002-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00003-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00004-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00005-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00006-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00007-of-00007.safetensors\"], \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/models_t5_umt5-xxl-enc-bf16.pth\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/Wan2.1_VAE.pth\"]",
+    "--learning_rate",
+    "5e-05",
+    "--num_epochs",
+    "1",
+    "--remove_prefix_in_ckpt",
+    "pipe.vace.",
+    "--output_path",
+    "/mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/low_high_cost_8gpu_latent_text_continue_si/real_train_output",
+    "--lora_base_model",
+    "vace",
+    "--lora_target_modules",
+    "q,k,v,o,ffn.0,ffn.2",
+    "--lora_rank",
+    "128",
+    "--extra_inputs",
+    "vace_video",
+    "--gradient_accumulation_steps",
+    "1",
+    "--dataset_num_workers",
+    "0",
+    "--weight_decay",
+    "0.01",
+    "--save_steps",
+    "5000",
+    "--lora_checkpoint",
+    "/mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/base_sft_rebuild_merged_8gpu_latent_text_cuda_init/real_train_output/step-15214.safetensors",
+    "--latent_memory_mode",
+    "text",
+    "--latent_memory_tokens",
+    "4",
+    "--latent_memory_hidden_dim",
+    "0",
+    "--latent_memory_scale",
+    "1.0",
+    "--latent_memory_init_std",
+    "0.02",
+    "--latent_memory_checkpoint",
+    "/mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/base_sft_rebuild_merged_8gpu_latent_text_cuda_init/real_train_output/step-15214.safetensors",
+    "--model_init_device",
+    "cuda"
+  ],
+  "command_pretty": "/opt/conda/bin/python3 -m accelerate.commands.launch --config_file /mnt/si002961ale4/default/lgy/shiying/low-high-new/scripts/training/edit-model/accelerate_config_8gpu_bf16.yaml --num_processes 8 /mnt/si002961ale4/default/lgy/shiying/low-high-new/Ditto/examples/wanvideo/model_training/train.py --dataset_base_path /mnt/si002961ale4/default/lgy/shiying/low-high-new --dataset_metadata_path /mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/low_high_cost_8gpu_latent_text_continue_si/real_train/metadata.train.csv --tokenizer_path /mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/google/umt5-xxl --data_file_keys video,vace_video --height 480 --width 832 --num_frames 17 --dataset_repeat 1 --model_paths '[[\"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00001-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00002-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00003-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00004-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00005-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00006-of-00007.safetensors\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/diffusion_pytorch_model-00007-of-00007.safetensors\"], \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/models_t5_umt5-xxl-enc-bf16.pth\", \"/mnt/si002961ale4/default/lgy/shiying/low-high-new/models/Wan2.1-VACE-14B/Wan2.1_VAE.pth\"]' --learning_rate 5e-05 --num_epochs 1 --remove_prefix_in_ckpt pipe.vace. --output_path /mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/low_high_cost_8gpu_latent_text_continue_si/real_train_output --lora_base_model vace --lora_target_modules q,k,v,o,ffn.0,ffn.2 --lora_rank 128 --extra_inputs vace_video --gradient_accumulation_steps 1 --dataset_num_workers 0 --weight_decay 0.01 --save_steps 5000 --lora_checkpoint /mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/base_sft_rebuild_merged_8gpu_latent_text_cuda_init/real_train_output/step-15214.safetensors --latent_memory_mode text --latent_memory_tokens 4 --latent_memory_hidden_dim 0 --latent_memory_scale 1.0 --latent_memory_init_std 0.02 --latent_memory_checkpoint /mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/base_sft_rebuild_merged_8gpu_latent_text_cuda_init/real_train_output/step-15214.safetensors --model_init_device cuda",
+  "train_metadata_path": "/mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/low_high_cost_8gpu_latent_text_continue_si/real_train/metadata.train.csv",
+  "val_metadata_path": "/mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/low_high_cost_8gpu_latent_text_continue_si/real_train/metadata.val.csv",
+  "output_path": "/mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/low_high_cost_8gpu_latent_text_continue_si/real_train_output",
+  "lora_checkpoint": "/mnt/si002961ale4/default/lgy/shiying/low-high-new/out/edit_model_face_stage1/base_sft_rebuild_merged_8gpu_latent_text_cuda_init/real_train_output/step-15214.safetensors",
+  "lora_rank": 128,
+  "models_root": "/mnt/si002961ale4/default/lgy/shiying/low-high-new/models",
+  "supported": true,
+  "status": "completed",
+  "log_path": "out/edit_model_face_stage1/low_high_cost_8gpu_latent_text_continue_si/real_train.log",
+  "returncode": 0
+}