training complete

README.md

@@ -26,16 +26,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.9344343517753922
+      value: 0.9326938978005622
     - name: Recall
       type: recall
-      value: 0.9522046449007069
+      value: 0.9491753618310333
     - name: F1
       type: f1
-      value: 0.9432358089522381
+      value: 0.9408624572524814
     - name: Accuracy
       type: accuracy
-      value: 0.9870783540354389
+      value: 0.9862394772473068
 ---
 
 <!-- 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: 0.0599
-- Precision: 0.9344
-- Recall: 0.9522
-- F1: 0.9432
-- Accuracy: 0.9871
+- Loss: 0.0612
+- Precision: 0.9327
+- Recall: 0.9492
+- F1: 0.9409
+- Accuracy: 0.9862
 
 ## Model description
 
@@ -80,9 +80,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.0752        | 1.0   | 1756 | 0.0615          | 0.9115    | 0.9391 | 0.9251 | 0.9830   |
-| 0.0343        | 2.0   | 3512 | 0.0617          | 0.9335    | 0.9492 | 0.9413 | 0.9861   |
-| 0.0203        | 3.0   | 5268 | 0.0599          | 0.9344    | 0.9522 | 0.9432 | 0.9871   |
+| 0.0763        | 1.0   | 1756 | 0.0688          | 0.9048    | 0.9325 | 0.9184 | 0.9809   |
+| 0.0345        | 2.0   | 3512 | 0.0641          | 0.9320    | 0.9458 | 0.9389 | 0.9854   |
+| 0.0219        | 3.0   | 5268 | 0.0612          | 0.9327    | 0.9492 | 0.9409 | 0.9862   |
 
 
 ### Framework versions