|
--- |
|
license: mit |
|
base_model: roberta-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: roberta-base-finetuned-ner |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# roberta-base-finetuned-ner |
|
|
|
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.1185 |
|
- Precision: 0.7791 |
|
- Recall: 0.8034 |
|
- F1: 0.7910 |
|
- Accuracy: 0.7680 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 6 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| No log | 0.19 | 50 | 1.2727 | 0.6213 | 0.4935 | 0.5501 | 0.4934 | |
|
| No log | 0.37 | 100 | 1.2623 | 0.6398 | 0.5312 | 0.5805 | 0.5263 | |
|
| No log | 0.56 | 150 | 1.2519 | 0.6609 | 0.5693 | 0.6117 | 0.5593 | |
|
| No log | 0.75 | 200 | 1.2423 | 0.6713 | 0.5940 | 0.6303 | 0.5815 | |
|
| No log | 0.93 | 250 | 1.2330 | 0.6828 | 0.6167 | 0.6481 | 0.6014 | |
|
| No log | 1.12 | 300 | 1.2241 | 0.6914 | 0.6388 | 0.6640 | 0.6219 | |
|
| No log | 1.31 | 350 | 1.2158 | 0.6962 | 0.6540 | 0.6744 | 0.6350 | |
|
| No log | 1.49 | 400 | 1.2076 | 0.6995 | 0.6637 | 0.6811 | 0.6434 | |
|
| No log | 1.68 | 450 | 1.2000 | 0.7048 | 0.6767 | 0.6905 | 0.6545 | |
|
| 1.2539 | 1.87 | 500 | 1.1926 | 0.7093 | 0.6880 | 0.6985 | 0.6645 | |
|
| 1.2539 | 2.05 | 550 | 1.1859 | 0.7148 | 0.6990 | 0.7068 | 0.6736 | |
|
| 1.2539 | 2.24 | 600 | 1.1793 | 0.7206 | 0.7092 | 0.7148 | 0.6824 | |
|
| 1.2539 | 2.43 | 650 | 1.1733 | 0.7269 | 0.7209 | 0.7239 | 0.6935 | |
|
| 1.2539 | 2.61 | 700 | 1.1676 | 0.7340 | 0.7306 | 0.7323 | 0.7025 | |
|
| 1.2539 | 2.8 | 750 | 1.1620 | 0.7385 | 0.7380 | 0.7382 | 0.7091 | |
|
| 1.2539 | 2.99 | 800 | 1.1569 | 0.7429 | 0.7451 | 0.744 | 0.7160 | |
|
| 1.2539 | 3.17 | 850 | 1.1521 | 0.7496 | 0.7560 | 0.7528 | 0.7265 | |
|
| 1.2539 | 3.36 | 900 | 1.1476 | 0.7539 | 0.7622 | 0.7580 | 0.7325 | |
|
| 1.2539 | 3.54 | 950 | 1.1435 | 0.7552 | 0.7657 | 0.7604 | 0.7349 | |
|
| 1.1751 | 3.73 | 1000 | 1.1399 | 0.7585 | 0.7718 | 0.7651 | 0.7405 | |
|
| 1.1751 | 3.92 | 1050 | 1.1364 | 0.7626 | 0.7789 | 0.7706 | 0.7470 | |
|
| 1.1751 | 4.1 | 1100 | 1.1332 | 0.7657 | 0.7835 | 0.7745 | 0.7513 | |
|
| 1.1751 | 4.29 | 1150 | 1.1303 | 0.7700 | 0.7895 | 0.7796 | 0.7561 | |
|
| 1.1751 | 4.48 | 1200 | 1.1278 | 0.7727 | 0.7934 | 0.7829 | 0.7589 | |
|
| 1.1751 | 4.66 | 1250 | 1.1256 | 0.7732 | 0.7945 | 0.7837 | 0.7600 | |
|
| 1.1751 | 4.85 | 1300 | 1.1237 | 0.7744 | 0.7960 | 0.7851 | 0.7614 | |
|
| 1.1751 | 5.04 | 1350 | 1.1221 | 0.7748 | 0.7973 | 0.7859 | 0.7622 | |
|
| 1.1751 | 5.22 | 1400 | 1.1208 | 0.7766 | 0.7995 | 0.7879 | 0.7643 | |
|
| 1.1751 | 5.41 | 1450 | 1.1198 | 0.7783 | 0.8021 | 0.7900 | 0.7665 | |
|
| 1.1363 | 5.6 | 1500 | 1.1191 | 0.7789 | 0.8032 | 0.7908 | 0.7675 | |
|
| 1.1363 | 5.78 | 1550 | 1.1187 | 0.7791 | 0.8034 | 0.7910 | 0.7680 | |
|
| 1.1363 | 5.97 | 1600 | 1.1185 | 0.7791 | 0.8034 | 0.7910 | 0.7680 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.19.0 |
|
- Tokenizers 0.15.2 |
|
|
