Instructions to use Angad23/upar2027 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- timm
How to use Angad23/upar2027 with timm:
import timm model = timm.create_model("hf-hub:Angad23/upar2027", pretrained=True) - Notebooks
- Google Colab
- Kaggle
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -13,7 +13,7 @@ Trained weights for our UPAR Challenge 2027 entries (WACV 2027 Real-World Survei
|
|
| 13 |
3rd place in Track 2 (attribute-based person retrieval, 69.11 mADM) and 79.44 in Track 1
|
| 14 |
(pedestrian attribute recognition).
|
| 15 |
|
| 16 |
-
Code, training recipe and instructions:
|
| 17 |
|
| 18 |
| File | Backbone (timm name) | Used in |
|
| 19 |
|---|---|---|
|
|
|
|
| 13 |
3rd place in Track 2 (attribute-based person retrieval, 69.11 mADM) and 79.44 in Track 1
|
| 14 |
(pedestrian attribute recognition).
|
| 15 |
|
| 16 |
+
Code, training recipe and instructions: https://github.com/angadbajwa23/upar2027
|
| 17 |
|
| 18 |
| File | Backbone (timm name) | Used in |
|
| 19 |
|---|---|---|
|