Training complete

README.md

@@ -25,16 +25,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.9323855182674823
+      value: 0.9135498687664042
     - name: Recall
       type: recall
-      value: 0.9491753618310333
+      value: 0.93722652305621
     - name: F1
       type: f1
-      value: 0.9407055291468601
+      value: 0.925236750290746
     - name: Accuracy
       type: accuracy
-      value: 0.9857685288750221
+      value: 0.9829869900512156
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -44,11 +44,11 @@ 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: 0.0639
-- Precision: 0.9324
-- Recall: 0.9492
-- F1: 0.9407
-- Accuracy: 0.9858
+- Loss: 0.0647
+- Precision: 0.9135
+- Recall: 0.9372
+- F1: 0.9252
+- Accuracy: 0.9830
 
 ## Model description
 
@@ -73,15 +73,13 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 1
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.0773        | 1.0   | 1756 | 0.0702          | 0.8949    | 0.9290 | 0.9116 | 0.9809   |
-| 0.0358        | 2.0   | 3512 | 0.0721          | 0.9315    | 0.9448 | 0.9381 | 0.9845   |
-| 0.0211        | 3.0   | 5268 | 0.0639          | 0.9324    | 0.9492 | 0.9407 | 0.9858   |
+| 0.0771        | 1.0   | 1756 | 0.0647          | 0.9135    | 0.9372 | 0.9252 | 0.9830   |
 
 
 ### Framework versions