Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -37,6 +37,23 @@ size_categories:
|
|
37 |
|
38 |
The **ImageNet-Hard** contains challenging images sourced from the validation datasets of ImageNet.
|
39 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
40 |
## Supported Tasks
|
41 |
|
42 |
- `image-classification`: The objective of this task is to classify an image into one or more classes, selected from 1000 ImageNet categories (allowing for multiple ground-truth labels per image).
|
|
|
37 |
|
38 |
The **ImageNet-Hard** contains challenging images sourced from the validation datasets of ImageNet.
|
39 |
|
40 |
+
|
41 |
+
### Classifiers Performance
|
42 |
+
|
43 |
+
|
44 |
+
| Model | Accuracy |
|
45 |
+
| ------------------- | ----- |
|
46 |
+
| ResNet-18 | 9.41 |
|
47 |
+
| ResNet-50 | 12.56 |
|
48 |
+
| ViT-B/32 | 15.95 |
|
49 |
+
| VGG19 | 10.32 |
|
50 |
+
| AlexNet | 6.35 |
|
51 |
+
| CLIP-ViT-L/14@224px | 1.86 |
|
52 |
+
| CLIP-ViT-L/14@336px | 2.02 |
|
53 |
+
| EfficientNet-L2-Ns | 34.23 |
|
54 |
+
|
55 |
+
|
56 |
+
|
57 |
## Supported Tasks
|
58 |
|
59 |
- `image-classification`: The objective of this task is to classify an image into one or more classes, selected from 1000 ImageNet categories (allowing for multiple ground-truth labels per image).
|