| { | |
| "dataset": { | |
| "repo_id": "hubnemo/so101_matchbox_reward_fpv_less_bias", | |
| "root": null, | |
| "episodes": null, | |
| "image_transforms": { | |
| "enable": true, | |
| "max_num_transforms": 3, | |
| "random_order": false, | |
| "tfs": { | |
| "color_jitter": { | |
| "weight": 1.0, | |
| "type": "ColorJitter", | |
| "kwargs": { | |
| "brightness": 0.2, | |
| "contrast": 0.2, | |
| "saturation": 0.2, | |
| "hue": 0.1 | |
| } | |
| } | |
| } | |
| }, | |
| "revision": null, | |
| "use_imagenet_stats": true, | |
| "video_backend": "torchcodec" | |
| }, | |
| "env": null, | |
| "policy": { | |
| "type": "reward_classifier", | |
| "n_obs_steps": 1, | |
| "normalization_mapping": { | |
| "VISUAL": "IDENTITY", | |
| "STATE": "MEAN_STD", | |
| "ACTION": "MEAN_STD" | |
| }, | |
| "input_features": { | |
| "observation.state": { | |
| "type": "STATE", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "observation.image.front": { | |
| "type": "VISUAL", | |
| "shape": [ | |
| 3, | |
| 480, | |
| 640 | |
| ] | |
| } | |
| }, | |
| "output_features": { | |
| "action": { | |
| "type": "ACTION", | |
| "shape": [ | |
| 6 | |
| ] | |
| } | |
| }, | |
| "device": "mps", | |
| "use_amp": false, | |
| "use_peft": false, | |
| "push_to_hub": true, | |
| "repo_id": "orellius/so101_matchbox_fpv_reward_model", | |
| "private": null, | |
| "tags": null, | |
| "license": null, | |
| "name": "reward_classifier", | |
| "num_classes": 2, | |
| "hidden_dim": 256, | |
| "latent_dim": 256, | |
| "image_embedding_pooling_dim": 8, | |
| "dropout_rate": 0.1, | |
| "model_name": "helper2424/resnet10", | |
| "model_type": "cnn", | |
| "num_cameras": 1, | |
| "learning_rate": 0.0003, | |
| "weight_decay": 0.01, | |
| "grad_clip_norm": 1.0 | |
| }, | |
| "output_dir": "outputs/train/2025-09-12/18-52-09_reward-classifier", | |
| "job_name": "reward-classifier", | |
| "resume": false, | |
| "seed": 2, | |
| "num_workers": 4, | |
| "batch_size": 16, | |
| "steps": 3000, | |
| "eval_freq": 1000, | |
| "log_freq": 100, | |
| "save_checkpoint": true, | |
| "save_freq": 10000, | |
| "use_validation": true, | |
| "val_split": 0.05, | |
| "val_freq": 500, | |
| "val_batch_size": 16, | |
| "use_policy_training_preset": true, | |
| "optimizer": { | |
| "type": "adamw", | |
| "lr": 0.0003, | |
| "weight_decay": 0.01, | |
| "grad_clip_norm": 1.0, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08 | |
| }, | |
| "scheduler": null, | |
| "eval": { | |
| "n_episodes": 50, | |
| "batch_size": 50, | |
| "use_async_envs": false | |
| }, | |
| "wandb": { | |
| "enable": true, | |
| "disable_artifact": false, | |
| "project": "reward-classifier", | |
| "entity": null, | |
| "notes": null, | |
| "run_id": "c62unon8", | |
| "mode": null | |
| }, | |
| "use_peft": false, | |
| "peft": { | |
| "target_modules": null, | |
| "modules_to_save": null, | |
| "method_type": "LORA", | |
| "init_type": null, | |
| "r": 16 | |
| } | |
| } |