| { | |
| "metadata": { | |
| "model_name": "SARM Progress Prediction", | |
| "description": "Stage-aware progress prediction model for robot manipulation tasks", | |
| "task": "clearing_food_from_table_into_fridge", | |
| "task_number": 25, | |
| "dataset": "IliaLarchenko/behavior_224_rgb", | |
| "version": "1.0", | |
| "author": "Your Name", | |
| "tags": ["robotics", "progress-estimation", "behavior-cloning"] | |
| }, | |
| "model": { | |
| "d_model": 768, | |
| "n_heads": 12, | |
| "n_layers": 8, | |
| "d_mlp": 512, | |
| "num_stages": 100, | |
| "d_state": 256, | |
| "num_tasks": 50 | |
| }, | |
| "training": { | |
| "max_steps": 10000, | |
| "learning_rate": 1e-4, | |
| "weight_decay": 1e-4, | |
| "batch_size": 16, | |
| "gradient_accumulation_steps": 4, | |
| "max_grad_norm": 1.0, | |
| "scheduler": "cosine", | |
| "stage_loss_weight": 1.0, | |
| "progress_loss_weight": 1.0, | |
| "validation_steps": 100, | |
| "save_steps": 200 | |
| }, | |
| "data": { | |
| "max_sequence_length": 13, | |
| "image_size": 224, | |
| "num_workers": 10, | |
| "val_workers": 10, | |
| "val_samples": 500, | |
| "train_episodes": [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], | |
| "val_episodes": [91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105], | |
| "seed": 42 | |
| }, | |
| "logging": { | |
| "project_name": "sarm-training", | |
| "run_name": null, | |
| "log_freq": 10, | |
| "checkpoint_dir": "checkpoints_sarm_25_2" | |
| } | |
| } |