| { | |
| "type": "smolvlafast", | |
| "n_obs_steps": 1, | |
| "input_features": { | |
| "observation.images.image": { | |
| "type": "VISUAL", | |
| "shape": [ | |
| 3, | |
| 256, | |
| 256 | |
| ] | |
| }, | |
| "observation.images.image2": { | |
| "type": "VISUAL", | |
| "shape": [ | |
| 3, | |
| 256, | |
| 256 | |
| ] | |
| }, | |
| "observation.state": { | |
| "type": "STATE", | |
| "shape": [ | |
| 8 | |
| ] | |
| } | |
| }, | |
| "output_features": { | |
| "action": { | |
| "type": "ACTION", | |
| "shape": [ | |
| 7 | |
| ] | |
| } | |
| }, | |
| "device": "cuda", | |
| "use_amp": false, | |
| "use_peft": false, | |
| "push_to_hub": true, | |
| "repo_id": "smolvla_fast_lr_1e-4_chunk1_max_act_dim7", | |
| "private": null, | |
| "tags": null, | |
| "license": null, | |
| "pretrained_path": null, | |
| "use_proprio": true, | |
| "chunk_size": 1, | |
| "n_action_steps": 1, | |
| "normalization_mapping": { | |
| "VISUAL": "IDENTITY", | |
| "STATE": "MEAN_STD", | |
| "ACTION": "MEAN_STD" | |
| }, | |
| "max_state_dim": 32, | |
| "max_action_dim": 7, | |
| "resize_imgs_with_padding": [ | |
| 512, | |
| 512 | |
| ], | |
| "empty_cameras": 0, | |
| "adapt_to_pi_aloha": false, | |
| "use_delta_joint_actions_aloha": false, | |
| "tokenizer_max_length": 128, | |
| "num_steps": 10, | |
| "use_cache": true, | |
| "freeze_vision_encoder": true, | |
| "train_expert_only": true, | |
| "train_state_proj": true, | |
| "optimizer_lr": 0.0001, | |
| "optimizer_betas": [ | |
| 0.9, | |
| 0.95 | |
| ], | |
| "optimizer_eps": 1e-08, | |
| "optimizer_weight_decay": 1e-10, | |
| "optimizer_grad_clip_norm": 10, | |
| "scheduler_warmup_steps": 1000, | |
| "scheduler_decay_steps": 100000, | |
| "scheduler_decay_lr": 2.5e-06, | |
| "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct", | |
| "load_vlm_weights": true, | |
| "vlm_model_dtype": "bfloat16", | |
| "add_image_special_tokens": false, | |
| "attention_mode": "cross_attn", | |
| "prefix_length": -1, | |
| "pad_language_to": "longest", | |
| "num_vlm_layers": -1, | |
| "min_period": 0.004, | |
| "max_period": 4.0, | |
| "fast_skip_tokens": 128, | |
| "padding_side": "right", | |
| "max_input_seq_len": 512, | |
| "max_decoding_steps": 128, | |
| "relaxed_action_decoding": true | |
| } |