Upload config.json with huggingface_hub
Browse files- config.json +47 -7
config.json
CHANGED
|
@@ -1,17 +1,57 @@
|
|
| 1 |
{
|
| 2 |
-
"model_type": "act",
|
| 3 |
-
"library_name": "lerobot",
|
| 4 |
"type": "act",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
"vision_backbone": "resnet18",
|
|
|
|
|
|
|
|
|
|
| 6 |
"dim_model": 512,
|
| 7 |
-
"dim_feedforward": 3200,
|
| 8 |
"n_heads": 8,
|
|
|
|
|
|
|
| 9 |
"n_encoder_layers": 4,
|
| 10 |
"n_decoder_layers": 1,
|
| 11 |
-
"
|
| 12 |
-
"n_action_steps": 100,
|
| 13 |
"latent_dim": 32,
|
|
|
|
| 14 |
"dropout": 0.1,
|
| 15 |
-
"
|
| 16 |
-
"created_at": "2025-11-11T21:44:13.314974Z"
|
| 17 |
}
|
|
|
|
| 1 |
{
|
|
|
|
|
|
|
| 2 |
"type": "act",
|
| 3 |
+
"n_obs_steps": 1,
|
| 4 |
+
"input_features": {
|
| 5 |
+
"observation.state": {
|
| 6 |
+
"type": "STATE",
|
| 7 |
+
"shape": [
|
| 8 |
+
6
|
| 9 |
+
]
|
| 10 |
+
},
|
| 11 |
+
"observation.images.camera1": {
|
| 12 |
+
"type": "VISUAL",
|
| 13 |
+
"shape": [
|
| 14 |
+
3,
|
| 15 |
+
240,
|
| 16 |
+
320
|
| 17 |
+
]
|
| 18 |
+
},
|
| 19 |
+
"observation.images.camera2": {
|
| 20 |
+
"type": "VISUAL",
|
| 21 |
+
"shape": [
|
| 22 |
+
3,
|
| 23 |
+
240,
|
| 24 |
+
320
|
| 25 |
+
]
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"output_features": {
|
| 29 |
+
"action": {
|
| 30 |
+
"type": "ACTION",
|
| 31 |
+
"shape": [
|
| 32 |
+
6
|
| 33 |
+
]
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"n_action_steps": 100,
|
| 37 |
+
"normalization_mapping": {
|
| 38 |
+
"VISUAL": "MEAN_STD",
|
| 39 |
+
"STATE": "MEAN_STD",
|
| 40 |
+
"ACTION": "MEAN_STD"
|
| 41 |
+
},
|
| 42 |
"vision_backbone": "resnet18",
|
| 43 |
+
"pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
|
| 44 |
+
"replace_final_stride_with_dilation": false,
|
| 45 |
+
"pre_norm": false,
|
| 46 |
"dim_model": 512,
|
|
|
|
| 47 |
"n_heads": 8,
|
| 48 |
+
"dim_feedforward": 3200,
|
| 49 |
+
"feedforward_activation": "relu",
|
| 50 |
"n_encoder_layers": 4,
|
| 51 |
"n_decoder_layers": 1,
|
| 52 |
+
"use_vae": true,
|
|
|
|
| 53 |
"latent_dim": 32,
|
| 54 |
+
"temporal_ensemble_coeff": null,
|
| 55 |
"dropout": 0.1,
|
| 56 |
+
"kl_weight": 10.0
|
|
|
|
| 57 |
}
|