Instructions to use lerobot/lingbot_va_libero_long with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use lerobot/lingbot_va_libero_long with LeRobot:
- Notebooks
- Google Colab
- Kaggle
Upload folder using huggingface_hub
Browse files- config.json +102 -0
- model.safetensors +3 -0
- policy_postprocessor.json +35 -0
- policy_preprocessor.json +57 -0
config.json
ADDED
|
@@ -0,0 +1,102 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"type": "lingbot_va",
|
| 3 |
+
"n_obs_steps": 1,
|
| 4 |
+
"input_features": {
|
| 5 |
+
"observation.images.image": {
|
| 6 |
+
"type": "VISUAL",
|
| 7 |
+
"shape": [
|
| 8 |
+
3,
|
| 9 |
+
256,
|
| 10 |
+
256
|
| 11 |
+
]
|
| 12 |
+
},
|
| 13 |
+
"observation.images.image2": {
|
| 14 |
+
"type": "VISUAL",
|
| 15 |
+
"shape": [
|
| 16 |
+
3,
|
| 17 |
+
256,
|
| 18 |
+
256
|
| 19 |
+
]
|
| 20 |
+
}
|
| 21 |
+
},
|
| 22 |
+
"output_features": {
|
| 23 |
+
"action": {
|
| 24 |
+
"type": "ACTION",
|
| 25 |
+
"shape": [
|
| 26 |
+
7
|
| 27 |
+
]
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"device": "cpu",
|
| 31 |
+
"use_amp": false,
|
| 32 |
+
"use_peft": false,
|
| 33 |
+
"push_to_hub": true,
|
| 34 |
+
"repo_id": null,
|
| 35 |
+
"private": null,
|
| 36 |
+
"tags": null,
|
| 37 |
+
"license": null,
|
| 38 |
+
"pretrained_path": null,
|
| 39 |
+
"patch_size": [
|
| 40 |
+
1,
|
| 41 |
+
2,
|
| 42 |
+
2
|
| 43 |
+
],
|
| 44 |
+
"num_attention_heads": 24,
|
| 45 |
+
"attention_head_dim": 128,
|
| 46 |
+
"in_channels": 48,
|
| 47 |
+
"out_channels": 48,
|
| 48 |
+
"action_dim": 30,
|
| 49 |
+
"text_dim": 4096,
|
| 50 |
+
"freq_dim": 256,
|
| 51 |
+
"ffn_dim": 14336,
|
| 52 |
+
"num_layers": 30,
|
| 53 |
+
"cross_attn_norm": true,
|
| 54 |
+
"eps": 1e-06,
|
| 55 |
+
"rope_max_seq_len": 1024,
|
| 56 |
+
"attn_mode": "torch",
|
| 57 |
+
"wan_pretrained_path": "robbyant/lingbot-va-posttrain-libero-long",
|
| 58 |
+
"dtype": "bfloat16",
|
| 59 |
+
"text_encoder_device": "cpu",
|
| 60 |
+
"obs_cam_keys": [
|
| 61 |
+
"observation.images.image",
|
| 62 |
+
"observation.images.image2"
|
| 63 |
+
],
|
| 64 |
+
"image_hflip": true,
|
| 65 |
+
"height": 128,
|
| 66 |
+
"width": 128,
|
| 67 |
+
"action_per_frame": 4,
|
| 68 |
+
"frame_chunk_size": 4,
|
| 69 |
+
"attn_window": 30,
|
| 70 |
+
"num_inference_steps": 20,
|
| 71 |
+
"video_exec_step": -1,
|
| 72 |
+
"action_num_inference_steps": 50,
|
| 73 |
+
"guidance_scale": 5.0,
|
| 74 |
+
"action_guidance_scale": 1.0,
|
| 75 |
+
"snr_shift": 5.0,
|
| 76 |
+
"action_snr_shift": 0.05,
|
| 77 |
+
"max_sequence_length": 512,
|
| 78 |
+
"used_action_channel_ids": [
|
| 79 |
+
0,
|
| 80 |
+
1,
|
| 81 |
+
2,
|
| 82 |
+
3,
|
| 83 |
+
4,
|
| 84 |
+
5,
|
| 85 |
+
6
|
| 86 |
+
],
|
| 87 |
+
"save_predicted_video": false,
|
| 88 |
+
"normalization_mapping": {
|
| 89 |
+
"VISUAL": "IDENTITY",
|
| 90 |
+
"STATE": "IDENTITY",
|
| 91 |
+
"ACTION": "IDENTITY"
|
| 92 |
+
},
|
| 93 |
+
"optimizer_lr": 1e-05,
|
| 94 |
+
"optimizer_betas": [
|
| 95 |
+
0.9,
|
| 96 |
+
0.95
|
| 97 |
+
],
|
| 98 |
+
"optimizer_eps": 1e-08,
|
| 99 |
+
"optimizer_weight_decay": 0.0001,
|
| 100 |
+
"optimizer_grad_clip_norm": 1.0,
|
| 101 |
+
"scheduler_warmup_steps": 1000
|
| 102 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dab54f38cb66ae1f49d0357e08ea74327ed39d081bc97f8abf2a1f5c5639d19a
|
| 3 |
+
size 10177841732
|
policy_postprocessor.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"name": "policy_postprocessor",
|
| 3 |
+
"steps": [
|
| 4 |
+
{
|
| 5 |
+
"registry_name": "lingbot_va_action_unnormalize",
|
| 6 |
+
"config": {
|
| 7 |
+
"action_q01": [
|
| 8 |
+
-0.6589285731315613,
|
| 9 |
+
-0.84375,
|
| 10 |
+
-0.9375,
|
| 11 |
+
-0.12107142806053162,
|
| 12 |
+
-0.15964286029338837,
|
| 13 |
+
-0.26571428775787354,
|
| 14 |
+
-1.0
|
| 15 |
+
],
|
| 16 |
+
"action_q99": [
|
| 17 |
+
0.8999999761581421,
|
| 18 |
+
0.8544642925262451,
|
| 19 |
+
0.9375,
|
| 20 |
+
0.17142857611179352,
|
| 21 |
+
0.1842857152223587,
|
| 22 |
+
0.34392857551574707,
|
| 23 |
+
1.0
|
| 24 |
+
]
|
| 25 |
+
}
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"registry_name": "device_processor",
|
| 29 |
+
"config": {
|
| 30 |
+
"device": "cpu",
|
| 31 |
+
"float_dtype": null
|
| 32 |
+
}
|
| 33 |
+
}
|
| 34 |
+
]
|
| 35 |
+
}
|
policy_preprocessor.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"name": "policy_preprocessor",
|
| 3 |
+
"steps": [
|
| 4 |
+
{
|
| 5 |
+
"registry_name": "rename_observations_processor",
|
| 6 |
+
"config": {
|
| 7 |
+
"rename_map": {}
|
| 8 |
+
}
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"registry_name": "to_batch_processor",
|
| 12 |
+
"config": {}
|
| 13 |
+
},
|
| 14 |
+
{
|
| 15 |
+
"registry_name": "normalizer_processor",
|
| 16 |
+
"config": {
|
| 17 |
+
"eps": 1e-08,
|
| 18 |
+
"features": {
|
| 19 |
+
"observation.images.image": {
|
| 20 |
+
"type": "VISUAL",
|
| 21 |
+
"shape": [
|
| 22 |
+
3,
|
| 23 |
+
256,
|
| 24 |
+
256
|
| 25 |
+
]
|
| 26 |
+
},
|
| 27 |
+
"observation.images.image2": {
|
| 28 |
+
"type": "VISUAL",
|
| 29 |
+
"shape": [
|
| 30 |
+
3,
|
| 31 |
+
256,
|
| 32 |
+
256
|
| 33 |
+
]
|
| 34 |
+
},
|
| 35 |
+
"action": {
|
| 36 |
+
"type": "ACTION",
|
| 37 |
+
"shape": [
|
| 38 |
+
7
|
| 39 |
+
]
|
| 40 |
+
}
|
| 41 |
+
},
|
| 42 |
+
"norm_map": {
|
| 43 |
+
"VISUAL": "IDENTITY",
|
| 44 |
+
"STATE": "IDENTITY",
|
| 45 |
+
"ACTION": "IDENTITY"
|
| 46 |
+
}
|
| 47 |
+
}
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"registry_name": "device_processor",
|
| 51 |
+
"config": {
|
| 52 |
+
"device": "cpu",
|
| 53 |
+
"float_dtype": null
|
| 54 |
+
}
|
| 55 |
+
}
|
| 56 |
+
]
|
| 57 |
+
}
|