DataLabHell/smolvla_so100_stack_cubes_v3

1{2    "type": "smolvla",3    "n_obs_steps": 1,4    "input_features": {5        "observation.state": {6            "type": "STATE",7            "shape": [8                69            ]10        },11        "observation.images.camera1": {12            "type": "VISUAL",13            "shape": [14                3,15                256,16                25617            ]18        },19        "observation.images.camera2": {20            "type": "VISUAL",21            "shape": [22                3,23                256,24                25625            ]26        },27        "observation.images.camera3": {28            "type": "VISUAL",29            "shape": [30                3,31                256,32                25633            ]34        }35    },36    "output_features": {37        "action": {38            "type": "ACTION",39            "shape": [40                641            ]42        }43    },44    "device": "cuda",45    "use_amp": false,46    "use_peft": false,47    "push_to_hub": true,48    "repo_id": "DataLabHell/smolvla_so100_stack_cubes_v3",49    "private": null,50    "tags": null,51    "license": null,52    "pretrained_path": "lerobot/smolvla_base",53    "pretrained_revision": null,54    "chunk_size": 50,55    "n_action_steps": 50,56    "normalization_mapping": {57        "VISUAL": "IDENTITY",58        "STATE": "MEAN_STD",59        "ACTION": "MEAN_STD"60    },61    "max_state_dim": 32,62    "max_action_dim": 32,63    "resize_imgs_with_padding": [64        512,65        51266    ],67    "empty_cameras": 0,68    "adapt_to_pi_aloha": false,69    "use_delta_joint_actions_aloha": false,70    "tokenizer_max_length": 48,71    "num_steps": 10,72    "use_cache": true,73    "freeze_vision_encoder": false,74    "train_expert_only": false,75    "train_state_proj": true,76    "optimizer_lr": 0.0001,77    "optimizer_betas": [78        0.9,79        0.9580    ],81    "optimizer_eps": 1e-08,82    "optimizer_weight_decay": 1e-10,83    "optimizer_grad_clip_norm": 10.0,84    "scheduler_warmup_steps": 1000,85    "scheduler_decay_steps": 30000,86    "scheduler_decay_lr": 2.5e-06,87    "vlm_model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",88    "load_vlm_weights": true,89    "add_image_special_tokens": false,90    "attention_mode": "cross_attn",91    "prefix_length": 0,92    "pad_language_to": "max_length",93    "num_expert_layers": 0,94    "num_vlm_layers": 16,95    "self_attn_every_n_layers": 2,96    "expert_width_multiplier": 0.75,97    "min_period": 0.004,98    "max_period": 4.0,99    "rtc_config": null,100    "compile_model": false,101    "compile_mode": "max-autotune"102}