PEFT
Safetensors
lora
vision-language
video
optical-flow
uav
drone
motion
embodied-ai
spatial-intelligence
Instructions to use choucsan/SIS-Motion with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use choucsan/SIS-Motion with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct") model = PeftModel.from_pretrained(base_model, "choucsan/SIS-Motion") - Notebooks
- Google Colab
- Kaggle
Upload sis_motion_config.json with huggingface_hub
Browse files- sis_motion_config.json +13 -0
sis_motion_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"format_version": 1,
|
| 3 |
+
"model_type": "motion-mllm",
|
| 4 |
+
"base_model_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
|
| 5 |
+
"connector_type": "visual_flow",
|
| 6 |
+
"connector_weights": "connector_weights.pt",
|
| 7 |
+
"videoflow_checkpoint": "MOF_kitti.pth",
|
| 8 |
+
"motion_config": {
|
| 9 |
+
"motion_dim": 128,
|
| 10 |
+
"down_ratio": 8,
|
| 11 |
+
"decoder_depth": 12
|
| 12 |
+
}
|
| 13 |
+
}
|