End of training

README.md

@@ -17,8 +17,8 @@ should 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 an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3486
-- Accuracy: 0.8832
+- Loss: 0.3052
+- Accuracy: 0.9113
 
 ## Model description
 
@@ -38,23 +38,70 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 16
-- eval_batch_size: 16
+- train_batch_size: 32
+- eval_batch_size: 32
 - seed: 42
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 64
+- total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 3
+- num_epochs: 50
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.4922        | 1.0   | 750  | 0.4845          | 0.8471   |
-| 0.3538        | 2.0   | 1500 | 0.3828          | 0.8711   |
-| 0.3324        | 3.0   | 2250 | 0.3486          | 0.8832   |
+| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|
+| 1.0436        | 1.0   | 375   | 1.0302          | 0.7701   |
+| 0.6703        | 2.0   | 750   | 0.6180          | 0.8448   |
+| 0.5315        | 3.0   | 1125  | 0.5132          | 0.8509   |
+| 0.4074        | 4.0   | 1500  | 0.4185          | 0.8671   |
+| 0.3998        | 5.0   | 1875  | 0.3854          | 0.8705   |
+| 0.3436        | 6.0   | 2250  | 0.3563          | 0.8778   |
+| 0.3501        | 7.0   | 2625  | 0.3156          | 0.888    |
+| 0.3284        | 8.0   | 3000  | 0.3145          | 0.8932   |
+| 0.3123        | 9.0   | 3375  | 0.3207          | 0.888    |
+| 0.2989        | 10.0  | 3750  | 0.3059          | 0.8939   |
+| 0.259         | 11.0  | 4125  | 0.2933          | 0.898    |
+| 0.2115        | 12.0  | 4500  | 0.3067          | 0.8931   |
+| 0.2928        | 13.0  | 4875  | 0.2869          | 0.8966   |
+| 0.2398        | 14.0  | 5250  | 0.2865          | 0.8989   |
+| 0.2187        | 15.0  | 5625  | 0.2955          | 0.9005   |
+| 0.2335        | 16.0  | 6000  | 0.2814          | 0.8998   |
+| 0.2165        | 17.0  | 6375  | 0.2863          | 0.8998   |
+| 0.2092        | 18.0  | 6750  | 0.2912          | 0.9022   |
+| 0.2002        | 19.0  | 7125  | 0.2769          | 0.9025   |
+| 0.163         | 20.0  | 7500  | 0.2906          | 0.9029   |
+| 0.1795        | 21.0  | 7875  | 0.2832          | 0.9065   |
+| 0.1568        | 22.0  | 8250  | 0.2908          | 0.8972   |
+| 0.1815        | 23.0  | 8625  | 0.2913          | 0.9055   |
+| 0.158         | 24.0  | 9000  | 0.2926          | 0.9057   |
+| 0.1672        | 25.0  | 9375  | 0.2810          | 0.9056   |
+| 0.1846        | 26.0  | 9750  | 0.2894          | 0.9032   |
+| 0.1599        | 27.0  | 10125 | 0.3073          | 0.9025   |
+| 0.1547        | 28.0  | 10500 | 0.2990          | 0.9045   |
+| 0.1342        | 29.0  | 10875 | 0.2938          | 0.9093   |
+| 0.1594        | 30.0  | 11250 | 0.2949          | 0.9058   |
+| 0.1582        | 31.0  | 11625 | 0.3076          | 0.9037   |
+| 0.1453        | 32.0  | 12000 | 0.2888          | 0.9086   |
+| 0.1643        | 33.0  | 12375 | 0.3031          | 0.9074   |
+| 0.1064        | 34.0  | 12750 | 0.3045          | 0.9046   |
+| 0.1661        | 35.0  | 13125 | 0.2968          | 0.909    |
+| 0.1345        | 36.0  | 13500 | 0.3027          | 0.9105   |
+| 0.1359        | 37.0  | 13875 | 0.3123          | 0.9069   |
+| 0.1261        | 38.0  | 14250 | 0.3079          | 0.9073   |
+| 0.1347        | 39.0  | 14625 | 0.3095          | 0.9095   |
+| 0.1364        | 40.0  | 15000 | 0.3020          | 0.9083   |
+| 0.108         | 41.0  | 15375 | 0.2934          | 0.9117   |
+| 0.1269        | 42.0  | 15750 | 0.3050          | 0.9125   |
+| 0.1187        | 43.0  | 16125 | 0.3144          | 0.9103   |
+| 0.11          | 44.0  | 16500 | 0.3073          | 0.9072   |
+| 0.113         | 45.0  | 16875 | 0.3125          | 0.9109   |
+| 0.0935        | 46.0  | 17250 | 0.3088          | 0.9129   |
+| 0.1287        | 47.0  | 17625 | 0.3085          | 0.9139   |
+| 0.1186        | 48.0  | 18000 | 0.3069          | 0.9118   |
+| 0.1353        | 49.0  | 18375 | 0.3205          | 0.9117   |
+| 0.1122        | 50.0  | 18750 | 0.3052          | 0.9113   |
 
 
 ### Framework versions