Model save

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1911
-- Accuracy: 1.0
+- Loss: 0.0630
+- Accuracy: 0.9852
 
 ## Model description
 
@@ -47,16 +47,21 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 5
+- num_epochs: 20
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Accuracy |
-|:-------------:|:------:|:----:|:---------------:|:--------:|
-| No log        | 0.8889 | 4    | 0.8267          | 0.9048   |
-| No log        | 2.0    | 9    | 0.4539          | 0.9683   |
-| 0.8656        | 2.8889 | 13   | 0.3035          | 0.9683   |
-| 0.8656        | 4.0    | 18   | 0.1911          | 1.0      |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 1.0732        | 1.0   | 5    | 0.9285          | 0.6667   |
+| 0.829         | 2.0   | 10   | 0.5274          | 0.8593   |
+| 0.5189        | 3.0   | 15   | 0.2473          | 0.9333   |
+| 0.3517        | 4.0   | 20   | 0.1405          | 0.9630   |
+| 0.2075        | 5.0   | 25   | 0.1797          | 0.9333   |
+| 0.1871        | 6.0   | 30   | 0.0936          | 0.9704   |
+| 0.1567        | 7.0   | 35   | 0.1113          | 0.9704   |
+| 0.1482        | 8.0   | 40   | 0.0791          | 0.9778   |
+| 0.1143        | 9.0   | 45   | 0.0630          | 0.9852   |
 
 
 ### Framework versions