metadata
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: bert-base-chinese-david-ner
results: []
bert-base-chinese-david-ner
This model is a fine-tuned version of bert-base-chinese on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0427
- Precision: 0.9485
- Recall: 0.9593
- F1: 0.9539
- Accuracy: 0.9908
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 4
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
1.0343 | 0.1 | 100 | 0.4009 | 0.3314 | 0.2922 | 0.3106 | 0.8495 |
0.2482 | 0.2 | 200 | 0.1478 | 0.7218 | 0.7878 | 0.7533 | 0.9524 |
0.1598 | 0.3 | 300 | 0.1212 | 0.7666 | 0.8386 | 0.8010 | 0.9585 |
0.1266 | 0.39 | 400 | 0.1038 | 0.7890 | 0.8602 | 0.8231 | 0.9680 |
0.1092 | 0.49 | 500 | 0.0863 | 0.8298 | 0.8856 | 0.8568 | 0.9733 |
0.0978 | 0.59 | 600 | 0.0912 | 0.8575 | 0.9022 | 0.8793 | 0.9730 |
0.1001 | 0.69 | 700 | 0.0675 | 0.8867 | 0.9047 | 0.8956 | 0.9802 |
0.09 | 0.79 | 800 | 0.0635 | 0.8993 | 0.9187 | 0.9089 | 0.9825 |
0.0817 | 0.89 | 900 | 0.0679 | 0.8849 | 0.9187 | 0.9015 | 0.9791 |
0.0786 | 0.99 | 1000 | 0.0572 | 0.8936 | 0.9288 | 0.9109 | 0.9829 |
0.0553 | 1.09 | 1100 | 0.0752 | 0.9 | 0.9263 | 0.9130 | 0.9802 |
0.0591 | 1.18 | 1200 | 0.0572 | 0.9059 | 0.9301 | 0.9179 | 0.9849 |
0.0382 | 1.28 | 1300 | 0.0598 | 0.9180 | 0.9390 | 0.9284 | 0.9864 |
0.0543 | 1.38 | 1400 | 0.0530 | 0.9274 | 0.9416 | 0.9344 | 0.9874 |
0.0519 | 1.48 | 1500 | 0.0558 | 0.9106 | 0.9314 | 0.9209 | 0.9854 |
0.0504 | 1.58 | 1600 | 0.0692 | 0.9100 | 0.9377 | 0.9237 | 0.9825 |
0.0426 | 1.68 | 1700 | 0.0535 | 0.9203 | 0.9390 | 0.9296 | 0.9865 |
0.0455 | 1.78 | 1800 | 0.0503 | 0.9313 | 0.9479 | 0.9395 | 0.9882 |
0.0477 | 1.88 | 1900 | 0.0446 | 0.9293 | 0.9517 | 0.9404 | 0.9883 |
0.0402 | 1.97 | 2000 | 0.0384 | 0.9437 | 0.9593 | 0.9515 | 0.9907 |
0.0342 | 2.07 | 2100 | 0.0462 | 0.9257 | 0.9492 | 0.9373 | 0.9887 |
0.021 | 2.17 | 2200 | 0.0546 | 0.9337 | 0.9492 | 0.9414 | 0.9882 |
0.0289 | 2.27 | 2300 | 0.0434 | 0.9424 | 0.9555 | 0.9489 | 0.9908 |
0.027 | 2.37 | 2400 | 0.0434 | 0.9353 | 0.9555 | 0.9453 | 0.9891 |
0.0231 | 2.47 | 2500 | 0.0427 | 0.9485 | 0.9593 | 0.9539 | 0.9908 |
0.0229 | 2.57 | 2600 | 0.0447 | 0.9352 | 0.9530 | 0.9440 | 0.9893 |
0.0251 | 2.67 | 2700 | 0.0448 | 0.9485 | 0.9593 | 0.9539 | 0.9900 |
0.0284 | 2.76 | 2800 | 0.0463 | 0.9423 | 0.9543 | 0.9482 | 0.9899 |
0.0244 | 2.86 | 2900 | 0.0449 | 0.9411 | 0.9543 | 0.9476 | 0.9893 |
0.0233 | 2.96 | 3000 | 0.0461 | 0.9411 | 0.9543 | 0.9476 | 0.9888 |
0.0148 | 3.06 | 3100 | 0.0461 | 0.9401 | 0.9581 | 0.9490 | 0.9900 |
0.0153 | 3.16 | 3200 | 0.0460 | 0.9388 | 0.9555 | 0.9471 | 0.9900 |
0.0147 | 3.26 | 3300 | 0.0466 | 0.9374 | 0.9517 | 0.9445 | 0.9894 |
0.0133 | 3.36 | 3400 | 0.0467 | 0.9385 | 0.9504 | 0.9444 | 0.9891 |
0.0217 | 3.46 | 3500 | 0.0457 | 0.9301 | 0.9466 | 0.9383 | 0.9892 |
0.0143 | 3.55 | 3600 | 0.0451 | 0.9363 | 0.9530 | 0.9446 | 0.9900 |
0.0077 | 3.65 | 3700 | 0.0466 | 0.9401 | 0.9568 | 0.9484 | 0.9906 |
0.0138 | 3.75 | 3800 | 0.0482 | 0.9401 | 0.9568 | 0.9484 | 0.9908 |
0.0168 | 3.85 | 3900 | 0.0486 | 0.9387 | 0.9543 | 0.9464 | 0.9894 |
0.0195 | 3.95 | 4000 | 0.0471 | 0.9387 | 0.9543 | 0.9464 | 0.9900 |
Framework versions
- Transformers 4.29.0.dev0
- Pytorch 1.10.1+cu113
- Datasets 2.11.0
- Tokenizers 0.13.3