Robotics
LeRobot
Safetensors
evo1
super_chatton_evo2 / config.json
imstevenpmwork's picture
Upload policy weights, train config and readme
039a209 verified
Raw
History Blame Contribute Delete
2.5 kB
{
"type": "evo1",
"n_obs_steps": 1,
"input_features": {
"observation.state": {
"type": "STATE",
"shape": [
6
]
},
"observation.images.front": {
"type": "VISUAL",
"shape": [
3,
480,
640
]
},
"observation.images.wrist": {
"type": "VISUAL",
"shape": [
3,
480,
640
]
}
},
"output_features": {
"action": {
"type": "ACTION",
"shape": [
6
]
}
},
"device": "cuda",
"use_amp": true,
"use_peft": false,
"push_to_hub": true,
"repo_id": "imstevenpmwork/super_chatton_evo2",
"private": null,
"tags": null,
"license": null,
"pretrained_path": null,
"training_stage": "stage2",
"chunk_size": 50,
"n_action_steps": 50,
"max_state_dim": 6,
"max_action_dim": 6,
"max_views": 3,
"image_resolution": [
448,
448
],
"empty_cameras": 0,
"postprocess_action_dim": null,
"binarize_gripper": false,
"gripper_index": 6,
"gripper_threshold": 0.5,
"gripper_below_threshold_value": 1.0,
"gripper_above_threshold_value": -1.0,
"normalization_mapping": {
"VISUAL": "IDENTITY",
"STATE": "MIN_MAX",
"ACTION": "MIN_MAX"
},
"vlm_model_name": "OpenGVLab/InternVL3-1B-hf",
"vlm_num_layers": 14,
"vlm_dtype": "bfloat16",
"use_flash_attn": true,
"action_head": "flowmatching",
"embed_dim": 896,
"hidden_dim": 1024,
"state_hidden_dim": 1024,
"num_heads": 8,
"num_layers": 8,
"dropout": 0.0,
"num_inference_timesteps": 32,
"num_categories": 1,
"return_cls_only": false,
"enable_gradient_checkpointing": true,
"gradient_checkpointing_use_reentrant": false,
"finetune_vlm": true,
"finetune_language_model": true,
"finetune_vision_model": true,
"finetune_action_head": true,
"apply_training_stage_defaults": true,
"task_field": "task",
"embodiment_id_field": null,
"default_embodiment_id": 0,
"optimizer_lr": 0.0001,
"optimizer_betas": [
0.9,
0.999
],
"optimizer_eps": 1e-08,
"optimizer_weight_decay": 1e-05,
"optimizer_grad_clip_norm": 1.0,
"scheduler_warmup_steps": 1000,
"drop_last": true
}