End of training

README.md

@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [FacebookAI/roberta-large](https://huggingface.co/FacebookAI/roberta-large) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6219
-- Accuracy: 0.9230
-- Precision: 0.4875
-- Recall: 0.6724
-- F1: 0.5652
+- Loss: 0.1962
+- Accuracy: 0.9127
+- Precision: 0.4457
+- Recall: 0.7069
+- F1: 0.5467
 
 ## Model description
 
@@ -44,28 +44,23 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 8.806006939763805e-06
+- learning_rate: 4.253164784470222e-05
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- num_epochs: 10
+- num_epochs: 5
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
-| 0.2713        | 1.0   | 486  | 0.3207          | 0.9153   | 0.4259    | 0.3966 | 0.4107 |
-| 0.2289        | 2.0   | 972  | 0.3147          | 0.8999   | 0.4038    | 0.7241 | 0.5185 |
-| 0.2022        | 3.0   | 1458 | 0.3260          | 0.9217   | 0.4819    | 0.6897 | 0.5674 |
-| 0.1713        | 4.0   | 1944 | 0.3697          | 0.9243   | 0.4928    | 0.5862 | 0.5354 |
-| 0.1295        | 5.0   | 2430 | 0.4309          | 0.9127   | 0.4390    | 0.6207 | 0.5143 |
-| 0.091         | 6.0   | 2916 | 0.5458          | 0.8960   | 0.3883    | 0.6897 | 0.4969 |
-| 0.0571        | 7.0   | 3402 | 0.5855          | 0.9191   | 0.4684    | 0.6379 | 0.5401 |
-| 0.0397        | 8.0   | 3888 | 0.5846          | 0.9153   | 0.4524    | 0.6552 | 0.5352 |
-| 0.0317        | 9.0   | 4374 | 0.5929          | 0.9191   | 0.4699    | 0.6724 | 0.5532 |
-| 0.0196        | 10.0  | 4860 | 0.6219          | 0.9230   | 0.4875    | 0.6724 | 0.5652 |
+| 0.3038        | 1.0   | 167  | 0.2109          | 0.9089   | 0.3898    | 0.3966 | 0.3932 |
+| 0.2729        | 2.0   | 334  | 0.2530          | 0.9012   | 0.4078    | 0.7241 | 0.5217 |
+| 0.243         | 3.0   | 501  | 0.2277          | 0.9114   | 0.4409    | 0.7069 | 0.5430 |
+| 0.2129        | 4.0   | 668  | 0.1612          | 0.9204   | 0.4767    | 0.7069 | 0.5694 |
+| 0.1673        | 5.0   | 835  | 0.1962          | 0.9127   | 0.4457    | 0.7069 | 0.5467 |
 
 
 ### Framework versions