训练完成

README.md

@@ -26,16 +26,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.9353555445052213
+      value: 0.9326827654647701
     - name: Recall
       type: recall
-      value: 0.9496802423426456
+      value: 0.9490070683271625
     - name: F1
       type: f1
-      value: 0.942463465553236
+      value: 0.9407741074407742
     - name: Accuracy
       type: accuracy
-      value: 0.9864602342968152
+      value: 0.986504385706717
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -45,10 +45,10 @@ 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: 0.0617
-- Precision: 0.9354
-- Recall: 0.9497
-- F1: 0.9425
+- Loss: 0.0595
+- Precision: 0.9327
+- Recall: 0.9490
+- F1: 0.9408
 - Accuracy: 0.9865
 
 ## Model description
@@ -80,14 +80,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.0749        | 1.0   | 1756 | 0.0611          | 0.9076    | 0.9360 | 0.9216 | 0.9826   |
-| 0.0349        | 2.0   | 3512 | 0.0623          | 0.9322    | 0.9470 | 0.9396 | 0.9856   |
-| 0.0208        | 3.0   | 5268 | 0.0617          | 0.9354    | 0.9497 | 0.9425 | 0.9865   |
+| 0.0774        | 1.0   | 1756 | 0.0701          | 0.9030    | 0.9302 | 0.9164 | 0.9808   |
+| 0.0366        | 2.0   | 3512 | 0.0641          | 0.9342    | 0.9461 | 0.9401 | 0.9855   |
+| 0.0225        | 3.0   | 5268 | 0.0595          | 0.9327    | 0.9490 | 0.9408 | 0.9865   |
 
 
 ### Framework versions
 
 - Transformers 4.47.0
-- Pytorch 2.5.1
-- Datasets 3.1.0
+- Pytorch 2.5.1+cu124
+- Datasets 3.2.0
 - Tokenizers 0.21.0