Upload 2 files
Browse files- config.json +35 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"visual_encoder": {
|
| 3 |
+
"image_encoder": "Dinov2Small",
|
| 4 |
+
"text_encoder": "t5-small",
|
| 5 |
+
"fusion_xformer": {
|
| 6 |
+
"num_layers": 3,
|
| 7 |
+
"d_model": 512,
|
| 8 |
+
"nhead": 8
|
| 9 |
+
},
|
| 10 |
+
"input_sensors": [
|
| 11 |
+
"raw_navigation_camera",
|
| 12 |
+
"raw_manipulation_camera",
|
| 13 |
+
"last_actions",
|
| 14 |
+
"an_object_is_in_hand"
|
| 15 |
+
],
|
| 16 |
+
"bbox_encoding_type": "positional"
|
| 17 |
+
},
|
| 18 |
+
"visual_text_encoder_class": "TextCondMultiCameraVisualEncoder",
|
| 19 |
+
"decoder": {
|
| 20 |
+
"num_layers": 3,
|
| 21 |
+
"d_model": 512,
|
| 22 |
+
"nhead": 8
|
| 23 |
+
},
|
| 24 |
+
"num_actions": 20,
|
| 25 |
+
"max_length": 1000,
|
| 26 |
+
"action_loss": true,
|
| 27 |
+
"use_llama_decoder": true,
|
| 28 |
+
"_attn_implementation_autoset": false,
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"architectures": [
|
| 31 |
+
"EarlyFusionCnnTransformer"
|
| 32 |
+
],
|
| 33 |
+
"model_type": "MM",
|
| 34 |
+
"transformers_version": "4.49.0"
|
| 35 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:baec9db1ad293692c7e11cf7dafaf7420273f508297a9e86a90740fe963b947b
|
| 3 |
+
size 321916536
|