Training complete

README.md

@@ -26,16 +26,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.9320885657633841
+      value: 0.9299867899603699
     - name: Recall
       type: recall
-      value: 0.9493436553349041
+      value: 0.9478290138000673
     - name: F1
       type: f1
-      value: 0.9406369851592463
+      value: 0.9388231371895316
     - name: Accuracy
       type: accuracy
-      value: 0.9859451345146288
+      value: 0.9857390946017542
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -45,11 +45,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the conll2003 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0621
-- Precision: 0.9321
-- Recall: 0.9493
-- F1: 0.9406
-- Accuracy: 0.9859
+- Loss: 0.0630
+- Precision: 0.9300
+- Recall: 0.9478
+- F1: 0.9388
+- Accuracy: 0.9857
 
 ## Model description
 
@@ -80,9 +80,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.0744        | 1.0   | 1756 | 0.0782          | 0.8921    | 0.9253 | 0.9084 | 0.9794   |
-| 0.0354        | 2.0   | 3512 | 0.0635          | 0.9294    | 0.9460 | 0.9376 | 0.9854   |
-| 0.021         | 3.0   | 5268 | 0.0621          | 0.9321    | 0.9493 | 0.9406 | 0.9859   |
+| 0.0767        | 1.0   | 1756 | 0.0671          | 0.9071    | 0.9334 | 0.9200 | 0.9812   |
+| 0.0333        | 2.0   | 3512 | 0.0760          | 0.9237    | 0.9394 | 0.9315 | 0.9836   |
+| 0.0223        | 3.0   | 5268 | 0.0630          | 0.9300    | 0.9478 | 0.9388 | 0.9857   |
 
 
 ### Framework versions