Update README.md
Browse files
README.md
CHANGED
@@ -2,25 +2,9 @@
|
|
2 |
library_name: keras
|
3 |
---
|
4 |
|
5 |
-
|
|
|
6 |
|
7 |
-
|
8 |
|
9 |
-
|
10 |
-
|
11 |
-
More information needed
|
12 |
-
|
13 |
-
## Training and evaluation data
|
14 |
-
|
15 |
-
More information needed
|
16 |
-
|
17 |
-
## Training Metrics
|
18 |
-
Model history needed
|
19 |
-
## Model Plot
|
20 |
-
|
21 |
-
<details>
|
22 |
-
<summary>View Model Plot</summary>
|
23 |
-
|
24 |
-
![Model Image](./model.png)
|
25 |
-
|
26 |
-
</details>
|
|
|
2 |
library_name: keras
|
3 |
---
|
4 |
|
5 |
+
This model is a TensorFlow port of DINO ViT B-16 [1]. The backbone of this model was pre-trained using the DINO pretext task. After that its head layer was trained
|
6 |
+
by keeping the backbone frozen. ImageNet-1k was used for training purposes. You can refer to [this notebook](https://github.com/sayakpaul/probing-vits/blob/main/notebooks/load-dino-weights-vitb16.ipynb) to know how the porting was done.
|
7 |
|
8 |
+
## References
|
9 |
|
10 |
+
[1] Emerging Properties in Self-Supervised Vision Transformers: https://arxiv.org/abs/2104.14294
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|