Upload LeWAM weights
Browse files- README.md +1 -0
- config.json +65 -15
- model.safetensors +2 -2
README.md
CHANGED
|
@@ -14,6 +14,7 @@ Joint video-action flow-matching model for robot control.
|
|
| 14 |
- **Context frames**: 32 @ 5 fps
|
| 15 |
- **Future frames**: 8
|
| 16 |
- **Action dim**: 6 @ 30 fps
|
|
|
|
| 17 |
|
| 18 |
## Usage
|
| 19 |
|
|
|
|
| 14 |
- **Context frames**: 32 @ 5 fps
|
| 15 |
- **Future frames**: 8
|
| 16 |
- **Action dim**: 6 @ 30 fps
|
| 17 |
+
- **Cameras**: image1, image2
|
| 18 |
|
| 19 |
## Usage
|
| 20 |
|
config.json
CHANGED
|
@@ -1,17 +1,67 @@
|
|
| 1 |
{
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"type": "lewam",
|
| 3 |
+
"n_obs_steps": 1,
|
| 4 |
+
"input_features": {
|
| 5 |
+
"observation.images.image1": {
|
| 6 |
+
"type": "VISUAL",
|
| 7 |
+
"shape": [
|
| 8 |
+
3,
|
| 9 |
+
480,
|
| 10 |
+
640
|
| 11 |
+
]
|
| 12 |
+
},
|
| 13 |
+
"observation.images.image2": {
|
| 14 |
+
"type": "VISUAL",
|
| 15 |
+
"shape": [
|
| 16 |
+
3,
|
| 17 |
+
480,
|
| 18 |
+
640
|
| 19 |
+
]
|
| 20 |
+
},
|
| 21 |
+
"observation.state": {
|
| 22 |
+
"type": "STATE",
|
| 23 |
+
"shape": [
|
| 24 |
+
6
|
| 25 |
+
]
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"output_features": {
|
| 29 |
+
"action": {
|
| 30 |
+
"type": "ACTION",
|
| 31 |
+
"shape": [
|
| 32 |
+
6
|
| 33 |
+
]
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"device": "cuda",
|
| 37 |
+
"use_amp": false,
|
| 38 |
+
"use_peft": false,
|
| 39 |
+
"push_to_hub": true,
|
| 40 |
+
"repo_id": null,
|
| 41 |
+
"private": null,
|
| 42 |
+
"tags": null,
|
| 43 |
+
"license": null,
|
| 44 |
+
"pretrained_path": null,
|
| 45 |
+
"model_dim": 512,
|
| 46 |
+
"depth": 12,
|
| 47 |
+
"num_heads": 8,
|
| 48 |
+
"mlp_ratio": 4.0,
|
| 49 |
+
"vlm_model_id": "HuggingFaceTB/SmolVLM2-256M-Video-Instruct",
|
| 50 |
+
"vlm_num_layers": 4,
|
| 51 |
+
"norm_strategy": "q2_q98",
|
| 52 |
+
"num_ode_steps": 10,
|
| 53 |
+
"smooth_actions": true,
|
| 54 |
+
"crop_size": 224,
|
| 55 |
+
"fps": 5,
|
| 56 |
+
"action_fps": 30,
|
| 57 |
+
"num_context_frames": 32,
|
| 58 |
+
"num_future_frames": 8,
|
| 59 |
+
"n_action_steps": 48,
|
| 60 |
+
"action_weight": 1.0,
|
| 61 |
+
"lang_drop_rate": 0.1,
|
| 62 |
+
"normalization_mapping": {
|
| 63 |
+
"VISUAL": "IDENTITY",
|
| 64 |
+
"STATE": "IDENTITY",
|
| 65 |
+
"ACTION": "IDENTITY"
|
| 66 |
+
}
|
| 67 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc8fa5a7451494a22ed9bd3f18262402c86dc6f55b568ccc6945d82296330a04
|
| 3 |
+
size 1280458528
|