|
--- |
|
base_model: haryoaw/scenario-TCR-NER_data-univner_half |
|
library_name: transformers |
|
license: mit |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: scenario-non-kd-po-ner-full-mdeberta_data-univner_half66 |
|
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. --> |
|
|
|
# scenario-non-kd-po-ner-full-mdeberta_data-univner_half66 |
|
|
|
This model is a fine-tuned version of [haryoaw/scenario-TCR-NER_data-univner_half](https://huggingface.co/haryoaw/scenario-TCR-NER_data-univner_half) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1199 |
|
- Precision: 0.8560 |
|
- Recall: 0.8660 |
|
- F1: 0.8609 |
|
- Accuracy: 0.9848 |
|
|
|
## 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: 3e-05 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- seed: 66 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 30 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 0.0064 | 0.5828 | 500 | 0.0946 | 0.8530 | 0.8696 | 0.8612 | 0.9847 | |
|
| 0.0072 | 1.1655 | 1000 | 0.0935 | 0.8563 | 0.8531 | 0.8547 | 0.9844 | |
|
| 0.0058 | 1.7483 | 1500 | 0.0977 | 0.8394 | 0.8530 | 0.8461 | 0.9836 | |
|
| 0.005 | 2.3310 | 2000 | 0.1050 | 0.8492 | 0.8609 | 0.8550 | 0.9840 | |
|
| 0.0054 | 2.9138 | 2500 | 0.1081 | 0.8503 | 0.8422 | 0.8462 | 0.9834 | |
|
| 0.0043 | 3.4965 | 3000 | 0.1210 | 0.8273 | 0.8775 | 0.8516 | 0.9830 | |
|
| 0.0049 | 4.0793 | 3500 | 0.1118 | 0.8413 | 0.8590 | 0.8501 | 0.9836 | |
|
| 0.0035 | 4.6620 | 4000 | 0.1137 | 0.8465 | 0.8647 | 0.8555 | 0.9837 | |
|
| 0.0031 | 5.2448 | 4500 | 0.1150 | 0.8430 | 0.8551 | 0.8490 | 0.9832 | |
|
| 0.0027 | 5.8275 | 5000 | 0.1169 | 0.8401 | 0.8590 | 0.8495 | 0.9836 | |
|
| 0.0027 | 6.4103 | 5500 | 0.1147 | 0.8517 | 0.8678 | 0.8597 | 0.9847 | |
|
| 0.0034 | 6.9930 | 6000 | 0.1163 | 0.8457 | 0.8651 | 0.8553 | 0.9842 | |
|
| 0.0024 | 7.5758 | 6500 | 0.1133 | 0.8523 | 0.8652 | 0.8587 | 0.9846 | |
|
| 0.0031 | 8.1585 | 7000 | 0.1170 | 0.8399 | 0.8577 | 0.8487 | 0.9834 | |
|
| 0.0019 | 8.7413 | 7500 | 0.1243 | 0.8413 | 0.8673 | 0.8541 | 0.9840 | |
|
| 0.0019 | 9.3240 | 8000 | 0.1230 | 0.8393 | 0.8726 | 0.8556 | 0.9841 | |
|
| 0.002 | 9.9068 | 8500 | 0.1218 | 0.8444 | 0.8549 | 0.8496 | 0.9839 | |
|
| 0.002 | 10.4895 | 9000 | 0.1205 | 0.8518 | 0.8651 | 0.8584 | 0.9846 | |
|
| 0.0017 | 11.0723 | 9500 | 0.1184 | 0.8643 | 0.8553 | 0.8598 | 0.9846 | |
|
| 0.0014 | 11.6550 | 10000 | 0.1316 | 0.8363 | 0.8717 | 0.8536 | 0.9838 | |
|
| 0.0016 | 12.2378 | 10500 | 0.1199 | 0.8560 | 0.8660 | 0.8609 | 0.9848 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.1.1+cu121 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.19.1 |
|
|
