update model card README.md

README.md

@@ -24,16 +24,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.9384946949602122
+      value: 0.929159802306425
     - name: Recall
       type: recall
-      value: 0.9527095254123191
+      value: 0.9491753618310333
     - name: F1
       type: f1
-      value: 0.9455486888257891
+      value: 0.939060939060939
     - name: Accuracy
       type: accuracy
-      value: 0.9865485371166186
+      value: 0.9861070230176017
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -43,11 +43,11 @@ 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: 0.9385
-- Recall: 0.9527
-- F1: 0.9455
-- Accuracy: 0.9865
+- Loss: 0.0634
+- Precision: 0.9292
+- Recall: 0.9492
+- F1: 0.9391
+- Accuracy: 0.9861
 
 ## Model description
 
@@ -78,9 +78,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.0903        | 1.0   | 1756 | 0.0760          | 0.9157    | 0.9354 | 0.9254 | 0.9809   |
-| 0.0354        | 2.0   | 3512 | 0.0612          | 0.9343    | 0.9500 | 0.9421 | 0.9863   |
-| 0.019         | 3.0   | 5268 | 0.0621          | 0.9385    | 0.9527 | 0.9455 | 0.9865   |
+| 0.0853        | 1.0   | 1756 | 0.0672          | 0.9092    | 0.9354 | 0.9221 | 0.9820   |
+| 0.0366        | 2.0   | 3512 | 0.0642          | 0.9308    | 0.9490 | 0.9398 | 0.9859   |
+| 0.0182        | 3.0   | 5268 | 0.0634          | 0.9292    | 0.9492 | 0.9391 | 0.9861   |
 
 
 ### Framework versions