Instructions to use maximellerbach/omx_multicubes_molmoact2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use maximellerbach/omx_multicubes_molmoact2 with LeRobot:
- Notebooks
- Google Colab
- Kaggle
File size: 3,601 Bytes
736ea5e | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 | {
"type": "molmoact2",
"n_obs_steps": 1,
"input_features": {
"observation.state": {
"type": "STATE",
"shape": [
6
]
},
"observation.images.wrist": {
"type": "VISUAL",
"shape": [
3,
480,
640
]
},
"observation.images.top": {
"type": "VISUAL",
"shape": [
3,
480,
640
]
}
},
"output_features": {
"action": {
"type": "ACTION",
"shape": [
6
]
}
},
"device": "cuda",
"use_amp": false,
"use_peft": false,
"push_to_hub": true,
"repo_id": "maximellerbach/omx_multicubes_molmoact2",
"private": null,
"tags": null,
"license": null,
"pretrained_path": null,
"checkpoint_path": "allenai/MolmoAct2",
"checkpoint_revision": null,
"checkpoint_force_download": false,
"chunk_size": 30,
"n_action_steps": 30,
"action_mode": "both",
"inference_action_mode": null,
"discrete_action_tokenizer": "allenai/MolmoAct2-FAST-Tokenizer",
"discrete_generation_max_steps": null,
"norm_tag": null,
"setup_type": "single omx robotic arm",
"control_mode": "joint position control",
"image_keys": [
"observation.images.top",
"observation.images.wrist"
],
"normalize_language": true,
"add_setup_tokens": true,
"add_control_tokens": true,
"normalize_gripper": true,
"num_state_tokens": 256,
"max_sequence_length": null,
"expected_max_action_dim": 32,
"num_flow_timesteps": 8,
"flow_matching_cutoff": 1.0,
"flow_matching_time_offset": 0.001,
"flow_matching_time_scale": 0.999,
"flow_matching_beta_alpha": 1.0,
"flow_matching_beta_beta": 1.5,
"num_inference_steps": null,
"mask_action_dim_padding": true,
"enable_inference_cuda_graph": true,
"per_episode_seed": false,
"eval_seed": null,
"rtc_config": null,
"enable_lora_vlm": false,
"lora_rank": 64,
"lora_alpha": 16,
"lora_dropout": 0.05,
"lora_bias": "none",
"enable_lora_action_expert": false,
"enable_knowledge_insulation": false,
"freeze_embedding": true,
"train_action_expert_only": false,
"gradient_checkpointing": true,
"model_dtype": "bfloat16",
"softmax_auxiliary_loss": true,
"softmax_auxiliary_loss_scale": 0.0001,
"discrete_loss_token_weighting": "root_subsegments_root_tokens",
"optimizer_lr": 1e-05,
"optimizer_vit_lr": 5e-06,
"optimizer_connector_lr": 5e-06,
"optimizer_action_expert_lr": 5e-05,
"optimizer_betas": [
0.9,
0.95
],
"optimizer_eps": 1e-06,
"optimizer_weight_decay": 0.0,
"optimizer_grad_clip_norm": 1.0,
"scheduler_warmup_steps": 200,
"scheduler_decay_steps": null,
"scheduler_decay_lr": 1e-06,
"normalization_mapping": {
"VISUAL": "IDENTITY",
"STATE": "QUANTILES",
"ACTION": "QUANTILES"
},
"dataset_feature_names": {
"action": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos",
"gripper.pos"
],
"observation.state": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos",
"gripper.pos"
]
}
} |