End of training

README.md

@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2184
-- F1: 0.9369
-- Accuracy: 0.9544
-- Precision: 0.9395
-- Recall: 0.9343
+- Loss: 0.2661
+- F1: 0.9386
+- Accuracy: 0.9557
+- Precision: 0.9403
+- Recall: 0.9370
 
 ## Model description
 
@@ -44,7 +44,7 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 32
+- train_batch_size: 16
 - eval_batch_size: 8
 - seed: 100
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -55,10 +55,10 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step  | Validation Loss | F1     | Accuracy | Precision | Recall |
 |:-------------:|:-----:|:-----:|:---------------:|:------:|:--------:|:---------:|:------:|
-| 0.2241        | 1.0   | 2542  | 0.1977          | 0.8902 | 0.9193   | 0.8768    | 0.9040 |
-| 0.1488        | 2.0   | 5084  | 0.1627          | 0.9187 | 0.9418   | 0.9287    | 0.9090 |
-| 0.0994        | 3.0   | 7626  | 0.1759          | 0.9339 | 0.9524   | 0.9383    | 0.9297 |
-| 0.0687        | 4.0   | 10168 | 0.2099          | 0.9396 | 0.9563   | 0.9391    | 0.9401 |
+| 0.2173        | 1.0   | 5084  | 0.2008          | 0.8888 | 0.9212   | 0.9076    | 0.8708 |
+| 0.1684        | 2.0   | 10168 | 0.2277          | 0.9216 | 0.9436   | 0.9276    | 0.9156 |
+| 0.1212        | 3.0   | 15252 | 0.2288          | 0.9320 | 0.9503   | 0.9228    | 0.9414 |
+| 0.0683        | 4.0   | 20336 | 0.2502          | 0.9424 | 0.9584   | 0.9429    | 0.9419 |
 
 
 ### Framework versions