File size: 4,174 Bytes
12ccb27 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 |
{
"version": "0.2.0",
"configurations": [
{
"name": "Debug MDPO Training",
"type": "python",
"request": "launch",
"program": "${workspaceFolder}/scripts/train/debug_mdpo_train.sh",
"console": "integratedTerminal",
"cwd": "${workspaceFolder}",
"env": {
"PYTHONPATH": "${workspaceFolder}:${env:PYTHONPATH}",
"CUDA_VISIBLE_DEVICES": "1",
"TRANSFORMERS_OFFLINE": "1",
"WANDB_PROJECT": "vtimellm"
},
"args": [],
"justMyCode": false,
"stopOnEntry": false
},
{
"name": "Debug Python Training Direct",
"type": "python",
"request": "launch",
"module": "vtimellm.train.train_dpo_mem",
"console": "integratedTerminal",
"cwd": "${workspaceFolder}",
"env": {
"PYTHONPATH": "${workspaceFolder}:${env:PYTHONPATH}",
"CUDA_VISIBLE_DEVICES": "1",
"TRANSFORMERS_OFFLINE": "1",
"WANDB_PROJECT": "vtimellm",
"RANK": "1",
"MASTER_PORT": "29571"
},
"args": [
"--deepspeed", "./scripts/zero2.json",
"--lora_enable", "True",
"--lora_r", "8",
"--lora_alpha", "128",
"--training_stage", "3",
"--finetuning", "True",
"--model_name_or_path", "./checkpoints/vicuna-7b-v1.5",
"--version", "v1",
"--data_path", "./data/activitynet/mdpo-train.json",
"--data_folder", "./data/activitynet/videos/train",
"--feat_folder", "./data/activitynet/clipvitl14-vtimellm.pth",
"--pretrain_mm_mlp_adapter", "./checkpoints/vtimellm-vicuna-v1-5-7b-stage1/mm_projector.bin",
"--stage2_path", "./checkpoints/vtimellm-vicuna-v1-5-7b-stage2",
"--stage3_path", "./checkpoints/vtimellm-vicuna-v1-5-7b-stage3",
"--stage4_path", "checkpoints/vtimellm-vicuna-v1-5-7b-activitynet-stage4",
"--output_dir", "./outputs/vtimellm-vicuna-v1-5-7b-activitynet-stage5",
"--bf16", "True",
"--max_steps", "100",
"--per_device_train_batch_size", "2",
"--gradient_accumulation_steps", "4",
"--evaluation_strategy", "no",
"--save_strategy", "no",
"--save_steps", "50000",
"--save_total_limit", "10",
"--learning_rate", "1e-6",
"--freeze_mm_mlp_adapter", "True",
"--weight_decay", "0.",
"--warmup_ratio", "0.1",
"--lr_scheduler_type", "cosine",
"--logging_steps", "1",
"--tf32", "True",
"--model_max_length", "2048",
"--gradient_checkpointing", "True",
"--dataloader_num_workers", "4",
"--lazy_preprocess", "True",
"--report_to", "none",
"--run_name", "vtimellm-vicuna-v1-5-7b-activitynet-stage5",
"--gamma", "0.0",
"--beta", "0.5",
"--dpo_alpha", "1.0",
"--train4dpo"
],
"justMyCode": false,
"stopOnEntry": false
},
{
"name": "Debug with DeepSpeed",
"type": "python",
"request": "launch",
"program": "${workspaceFolder}/scripts/train/debug_deepspeed.py",
"console": "integratedTerminal",
"cwd": "${workspaceFolder}",
"env": {
"PYTHONPATH": "${workspaceFolder}:${env:PYTHONPATH}",
"CUDA_VISIBLE_DEVICES": "1",
"TRANSFORMERS_OFFLINE": "1",
"WANDB_PROJECT": "vtimellm",
"RANK": "1",
"MASTER_PORT": "29571"
},
"args": [],
"justMyCode": false,
"stopOnEntry": false
}
]
}
|