Training complete

README.md

@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/biogpt](https://huggingface.co/microsoft/biogpt) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1491
-- Precision: 0.4416
-- Recall: 0.5578
-- F1: 0.4930
-- Accuracy: 0.9576
+- Loss: 0.0922
+- Precision: 0.6758
+- Recall: 0.7814
+- F1: 0.7248
+- Accuracy: 0.9791
 
 ## Model description
 
@@ -55,9 +55,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.3004        | 1.0   | 679  | 0.1688          | 0.3313    | 0.4244 | 0.3721 | 0.9476   |
-| 0.1682        | 2.0   | 1358 | 0.1706          | 0.3805    | 0.5667 | 0.4553 | 0.9503   |
-| 0.0935        | 3.0   | 2037 | 0.1491          | 0.4416    | 0.5578 | 0.4930 | 0.9576   |
+| 0.1026        | 1.0   | 679  | 0.0663          | 0.6242    | 0.7853 | 0.6956 | 0.9772   |
+| 0.0487        | 2.0   | 1358 | 0.0710          | 0.6842    | 0.8094 | 0.7416 | 0.9789   |
+| 0.014         | 3.0   | 2037 | 0.0922          | 0.6758    | 0.7814 | 0.7248 | 0.9791   |
 
 
 ### Framework versions