End of training

README.md

@@ -19,10 +19,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google-bert/bert-base-uncased](https://huggingface.co/google-bert/bert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3626
-- Precision: 0.8862
-- Recall: 0.8829
-- F1: 0.8845
+- Loss: 0.3184
+- Precision: 0.8894
+- Recall: 0.8897
+- F1: 0.8894
 
 ## Model description
 
@@ -42,11 +42,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 8
-- eval_batch_size: 8
+- train_batch_size: 16
+- eval_batch_size: 16
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 16
+- total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
@@ -56,9 +56,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss | Precision | Recall | F1     |
 |:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|
-| 0.2839        | 0.9995 | 1024 | 0.2913          | 0.8693    | 0.8853 | 0.8663 |
-| 0.2686        | 2.0    | 2049 | 0.2956          | 0.8914    | 0.8804 | 0.8849 |
-| 0.2786        | 2.9985 | 3072 | 0.3626          | 0.8862    | 0.8829 | 0.8845 |
+| 0.3301        | 0.9990 | 512  | 0.3780          | 0.8509    | 0.8697 | 0.8514 |
+| 0.3103        | 2.0    | 1025 | 0.2916          | 0.8907    | 0.8848 | 0.8870 |
+| 0.1607        | 2.9971 | 1536 | 0.3184          | 0.8894    | 0.8897 | 0.8894 |
 
 
 ### Framework versions