zhiguoxu
/

bert-base-chinese-finetuned-split

@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [bert-base-chinese](https://huggingface.co/bert-base-chinese) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0141
 - F1: 1.0
 ## Model description
@@ -36,8 +36,8 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 20
-- eval_batch_size: 20
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
@@ -47,26 +47,26 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 3.4933        | 1.0   | 1    | 1.6514          | 0.0    |
-| 1.7503        | 2.0   | 2    | 0.8328          | 0.0    |
-| 0.888         | 3.0   | 3    | 0.5791          | 0.0    |
-| 0.5987        | 4.0   | 4    | 0.4557          | 0.0    |
-| 0.4704        | 5.0   | 5    | 0.3322          | 0.0    |
-| 0.3619        | 6.0   | 6    | 0.2705          | 0.0    |
-| 0.3121        | 7.0   | 7    | 0.2379          | 0.5161 |
-| 0.2764        | 8.0   | 8    | 0.1927          | 0.6471 |
-| 0.2232        | 9.0   | 9    | 0.1627          | 0.6207 |
-| 0.2012        | 10.0  | 10   | 0.1334          | 0.7879 |
-| 0.1483        | 11.0  | 11   | 0.1011          | 0.8333 |
-| 0.1381        | 12.0  | 12   | 0.0770          | 0.9500 |
-| 0.1037        | 13.0  | 13   | 0.0602          | 0.9756 |
-| 0.0973        | 14.0  | 14   | 0.0429          | 1.0    |
-| 0.0728        | 15.0  | 15   | 0.0310          | 1.0    |
-| 0.0541        | 16.0  | 16   | 0.0241          | 1.0    |
-| 0.0464        | 17.0  | 17   | 0.0194          | 1.0    |
-| 0.0369        | 18.0  | 18   | 0.0165          | 1.0    |
-| 0.031         | 19.0  | 19   | 0.0149          | 1.0    |
-| 0.0303        | 20.0  | 20   | 0.0141          | 1.0    |
 ### Framework versions

 This model is a fine-tuned version of [bert-base-chinese](https://huggingface.co/bert-base-chinese) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.0248
 - F1: 1.0
 ## Model description
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
+- train_batch_size: 22
+- eval_batch_size: 22
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 | Training Loss | Epoch | Step | Validation Loss | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
+| 4.227         | 1.0   | 1    | 2.1377          | 0.0    |
+| 2.2277        | 2.0   | 2    | 0.8621          | 0.0    |
+| 0.9409        | 3.0   | 3    | 0.5773          | 0.0    |
+| 0.5981        | 4.0   | 4    | 0.4644          | 0.0    |
+| 0.4765        | 5.0   | 5    | 0.3439          | 0.0    |
+| 0.3635        | 6.0   | 6    | 0.3140          | 0.0    |
+| 0.349         | 7.0   | 7    | 0.2688          | 0.0    |
+| 0.3031        | 8.0   | 8    | 0.2333          | 0.0    |
+| 0.2643        | 9.0   | 9    | 0.2003          | 0.0870 |
+| 0.2364        | 10.0  | 10   | 0.1592          | 0.5333 |
+| 0.1909        | 11.0  | 11   | 0.1369          | 0.8571 |
+| 0.1674        | 12.0  | 12   | 0.1230          | 0.8980 |
+| 0.1414        | 13.0  | 13   | 0.0983          | 0.9362 |
+| 0.1274        | 14.0  | 14   | 0.0718          | 0.9778 |
+| 0.0944        | 15.0  | 15   | 0.0531          | 1.0    |
+| 0.0833        | 16.0  | 16   | 0.0405          | 1.0    |
+| 0.0727        | 17.0  | 17   | 0.0328          | 1.0    |
+| 0.0557        | 18.0  | 18   | 0.0284          | 1.0    |
+| 0.0514        | 19.0  | 19   | 0.0259          | 1.0    |
+| 0.0456        | 20.0  | 20   | 0.0248          | 1.0    |
 ### Framework versions