Add results table
Browse files
README.md
CHANGED
|
@@ -73,7 +73,16 @@ The model was trained on a single 8-GPU node for 3 days. Training resolution is
|
|
| 73 |
|
| 74 |
## Evaluation results
|
| 75 |
|
| 76 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 77 |
|
| 78 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
|
| 79 |
|
|
|
|
| 73 |
|
| 74 |
## Evaluation results
|
| 75 |
|
| 76 |
+
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|
| 77 |
+
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
|
| 78 |
+
| DeiT-tiny | 72.2 | 91.1 | 5M | https://huggingface.co/facebook/deit-tiny-patch16-224 |
|
| 79 |
+
| DeiT-small | 79.9 | 95.0 | 22M | https://huggingface.co/facebook/deit-small-patch16-224 |
|
| 80 |
+
| **DeiT-base** | **81.8** | **95.6** | **86M** | **https://huggingface.co/facebook/deit-base-patch16-224** |
|
| 81 |
+
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
|
| 82 |
+
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
|
| 83 |
+
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
|
| 84 |
+
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
|
| 85 |
+
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
|
| 86 |
|
| 87 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
|
| 88 |
|