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 [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7975
-- Matthews Correlation: 0.5282
+- Loss: 0.9412
+- Matthews Correlation: 0.5101
 
 ## Model description
 
@@ -37,23 +37,21 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 16
+- learning_rate: 1.2157694831030876e-05
+- train_batch_size: 4
 - eval_batch_size: 16
-- seed: 42
+- seed: 38
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 5
+- num_epochs: 3
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
 |:-------------:|:-----:|:----:|:---------------:|:--------------------:|
-| 0.5187        | 1.0   | 535  | 0.4530          | 0.4578               |
-| 0.3439        | 2.0   | 1070 | 0.4707          | 0.5192               |
-| 0.2335        | 3.0   | 1605 | 0.6058          | 0.5253               |
-| 0.1736        | 4.0   | 2140 | 0.7975          | 0.5282               |
-| 0.1294        | 5.0   | 2675 | 0.8228          | 0.5252               |
+| 0.5486        | 1.0   | 2138 | 0.5384          | 0.4738               |
+| 0.4633        | 2.0   | 4276 | 0.7257          | 0.4895               |
+| 0.3176        | 3.0   | 6414 | 0.9412          | 0.5101               |
 
 
 ### Framework versions