Image Classification
timm
PyTorch
bdd100k-toolkit
computer-vision
autonomous-driving
self-driving-cars
driving-scenes
weather-classification
Eval Results (legacy)
Instructions to use dronefreak/bdd100k-weather-efficientvit_b1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- timm
How to use dronefreak/bdd100k-weather-efficientvit_b1 with timm:
import timm model = timm.create_model("hf-hub:dronefreak/bdd100k-weather-efficientvit_b1", pretrained=True) - Notebooks
- Google Colab
- Kaggle
Merge dev into main
Browse files
README.md
CHANGED
|
@@ -146,7 +146,15 @@ All runs below were evaluated on the same `test` split, sorted by top-1 accuracy
|
|
| 146 |
|
| 147 |
| Model | Top-1 | Macro F1 | Balanced acc | Macro precision |
|
| 148 |
|---|---|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 149 |
| ConvNeXt-Atto | 83.00% | 67.44% | 65.25% | 81.38% |
|
|
|
|
| 150 |
| EfficientViT-B0 | 82.79% | 65.10% | 63.62% | 67.11% |
|
| 151 |
| **EfficientViT-B1** | **82.67%** | **64.95%** | **63.59%** | **66.83%** |
|
| 152 |
| MobileNetV4-Conv-Large | 82.20% | 66.62% | 64.68% | 80.26% |
|
|
|
|
| 146 |
|
| 147 |
| Model | Top-1 | Macro F1 | Balanced acc | Macro precision |
|
| 148 |
|---|---|---|---|---|
|
| 149 |
+
| TinyViT-21M | 83.90% | 68.65% | 67.15% | 74.81% |
|
| 150 |
+
| EfficientViT-B3 | 83.51% | 68.17% | 66.34% | 81.75% |
|
| 151 |
+
| EfficientViT-B2 | 83.46% | 66.07% | 65.04% | 67.54% |
|
| 152 |
+
| EfficientFormerV2-L | 83.27% | 65.93% | 65.03% | 67.12% |
|
| 153 |
+
| RepViT-M1.5 | 83.19% | 67.78% | 65.71% | 81.56% |
|
| 154 |
+
| EfficientViT-L1 | 83.13% | 65.70% | 65.11% | 66.46% |
|
| 155 |
+
| RepViT-M2.3 | 83.02% | 65.46% | 64.33% | 67.02% |
|
| 156 |
| ConvNeXt-Atto | 83.00% | 67.44% | 65.25% | 81.38% |
|
| 157 |
+
| EfficientFormerV2-S2 | 82.90% | 65.50% | 64.68% | 66.98% |
|
| 158 |
| EfficientViT-B0 | 82.79% | 65.10% | 63.62% | 67.11% |
|
| 159 |
| **EfficientViT-B1** | **82.67%** | **64.95%** | **63.59%** | **66.83%** |
|
| 160 |
| MobileNetV4-Conv-Large | 82.20% | 66.62% | 64.68% | 80.26% |
|