Training in progress epoch 3

README.md

@@ -19,11 +19,11 @@ probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the mnist dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.9608
-- Train Accuracy: 0.9302
-- Validation Loss: 0.7593
-- Validation Accuracy: 0.9302
-- Epoch: 2
+- Train Loss: 0.6974
+- Train Accuracy: 0.9448
+- Validation Loss: 0.5795
+- Validation Accuracy: 0.9448
+- Epoch: 3
 
 ## Model description
 
@@ -52,6 +52,7 @@ The following hyperparameters were used during training:
 | 2.0110     | 0.8531         | 1.5291          | 0.8531              | 0     |
 | 1.3596     | 0.9151         | 0.9948          | 0.9151              | 1     |
 | 0.9608     | 0.9302         | 0.7593          | 0.9302              | 2     |
+| 0.6974     | 0.9448         | 0.5795          | 0.9448              | 3     |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ace3abe90e9e76937f9d93d6634087661220f5ec1de1409db23e11a63a308c6
+oid sha256:44c252472d79b5c3d393706e93e2a344e49ed986ce7a25255759d09c617c6b72
 size 343494328