litevla-ms / train_config.json
ducido's picture
Initial commit
b3eaccf verified
{
"dataset": {
"repo_id": ".",
"root": "/pfss/mlde/workspaces/mlde_wsp_IAS_SAMMerge/VLA/LIBERO/merged_libero_scale_100_mask_depth_noops_lerobot",
"episodes": null,
"image_transforms": {
"enable": true,
"max_num_transforms": 3,
"random_order": false,
"image_tfs": {
"hue": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"hue": [
-0.05,
0.05
]
}
},
"contrast": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"contrast": [
0.8,
1.2
]
}
},
"sharpness": {
"weight": 1.0,
"type": "SharpnessJitter",
"kwargs": {
"sharpness": [
0.5,
1.5
]
}
},
"brightness": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"brightness": [
0.8,
1.2
]
}
},
"saturation": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"saturation": [
0.5,
1.5
]
}
},
"crop_resize": {
"weight": 1.0,
"type": "RandomResizedCrop",
"kwargs": {
"size": [
256,
256
],
"ratio": [
1,
1
],
"scale": [
0.9,
0.95
]
}
},
"rotate": {
"weight": 1.0,
"type": "RandomRotate",
"kwargs": {
"degrees": [
-5,
5
]
}
}
},
"wrist_tfs": {
"hue": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"hue": [
-0.05,
0.05
]
}
},
"contrast": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"contrast": [
0.8,
1.2
]
}
},
"sharpness": {
"weight": 1.0,
"type": "SharpnessJitter",
"kwargs": {
"sharpness": [
0.5,
1.5
]
}
},
"brightness": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"brightness": [
0.8,
1.2
]
}
},
"saturation": {
"weight": 1.0,
"type": "ColorJitter",
"kwargs": {
"saturation": [
0.5,
1.5
]
}
}
}
},
"revision": null,
"use_imagenet_stats": true,
"video_backend": "torchcodec",
"vqa_data_path": null
},
"env": null,
"policy": {
"type": "smolvla",
"n_obs_steps": 1,
"normalization_mapping": {
"VISUAL": "IDENTITY",
"STATE": "MEAN_STD",
"ACTION": "MEAN_STD"
},
"input_features": {
"observation.images.image": {
"type": "VISUAL",
"shape": [
3,
256,
256
]
},
"observation.images.wrist_image": {
"type": "VISUAL",
"shape": [
3,
256,
256
]
},
"observation.images.image_mask": {
"type": "VISUAL",
"shape": [
3,
256,
256
]
},
"observation.images.wrist_mask": {
"type": "VISUAL",
"shape": [
3,
256,
256
]
},
"observation.images.object_of_interest_mask": {
"type": "VISUAL",
"shape": [
3,
256,
256
]
},
"observation.images.object_of_interest_wrist_mask": {
"type": "VISUAL",
"shape": [
3,
256,
256
]
},
"observation.state": {
"type": "STATE",
"shape": [
8
]
},
"observation.states.ee_state": {
"type": "STATE",
"shape": [
6
]
},
"observation.states.joint_state": {
"type": "STATE",
"shape": [
7
]
},
"observation.states.gripper_state": {
"type": "STATE",
"shape": [
2
]
}
},
"output_features": {
"action": {
"type": "ACTION",
"shape": [
7
]
}
},
"device": "cuda",
"use_amp": false,
"gradient_accumulation_steps": 1,
"chunk_size": 50,
"n_action_steps": 1,
"max_state_dim": 32,
"max_action_dim": 32,
"resize_imgs_with_padding": [
512,
512
],
"empty_cameras": 0,
"adapt_to_pi_aloha": false,
"use_delta_joint_actions_aloha": false,
"tokenizer_max_length": 48,
"num_steps": 10,
"use_cache": true,
"freeze_vision_encoder": true,
"train_expert_only": false,
"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": 10.0,
"scheduler_warmup_steps": 1000,
"scheduler_decay_steps": 100000,
"scheduler_decay_lr": 2.5e-06,
"vlm_model_name": "/pfss/mlde/workspaces/mlde_wsp_IAS_SAMMerge/VLA/duc/VLA-Humanoid-MW/SmolVLM2-500M-Video-Instruct",
"load_vlm_weights": true,
"add_image_special_tokens": false,
"attention_mode": "cross_attn",
"prefix_length": 0,
"pad_language_to": "max_length",
"num_expert_layers": 0,
"num_vlm_layers": 16,
"self_attn_every_n_layers": 2,
"expert_width_multiplier": 0.75,
"min_period": 0.004,
"max_period": 4.0,
"of_path": "/pfss/mlde/workspaces/mlde_wsp_IAS_SAMMerge/VLA/duc/VLA-Humanoid-MW/libero_100%_of.h5"
},
"output_dir": "outputs/train/2026-01-04/07-20-22_libero_100%_scratch_vlo_v2_after_resi_half_layer_100kdecay",
"job_name": "libero_100%_scratch_vlo_v2_after_resi_half_layer_100kdecay",
"resume": true,
"seed": 42,
"num_workers": 8,
"batch_size": 64,
"steps": 100000,
"eval_freq": 20000,
"log_freq": 200,
"save_checkpoint": true,
"save_freq": 10000,
"use_policy_training_preset": true,
"optimizer": {
"type": "adamw",
"lr": 0.0001,
"weight_decay": 1e-10,
"grad_clip_norm": 10.0,
"betas": [
0.9,
0.95
],
"eps": 1e-08
},
"scheduler": {
"type": "cosine_decay_with_warmup",
"num_warmup_steps": 1000,
"num_decay_steps": 100000,
"peak_lr": 0.0001,
"decay_lr": 2.5e-06
},
"eval": {
"n_episodes": 50,
"batch_size": 50,
"use_async_envs": false
},
"wandb": {
"enable": true,
"disable_artifact": true,
"project": "smolvla",
"entity": "Robotics_VLA",
"notes": null,
"run_id": null,
"mode": "online"
}
}