|
--- |
|
base_model: haryoaw/scenario-TCR-NER_data-univner_full |
|
library_name: transformers |
|
license: mit |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: scenario-non-kd-scr-ner-full_data-univner_full55 |
|
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-scr-ner-full_data-univner_full55 |
|
|
|
This model is a fine-tuned version of [haryoaw/scenario-TCR-NER_data-univner_full](https://huggingface.co/haryoaw/scenario-TCR-NER_data-univner_full) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1084 |
|
- Precision: 0.8506 |
|
- Recall: 0.8628 |
|
- F1: 0.8567 |
|
- Accuracy: 0.9845 |
|
|
|
## 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: 55 |
|
- 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 | |
|
|:-------------:|:------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 0.0097 | 0.2910 | 500 | 0.0794 | 0.8535 | 0.8668 | 0.8601 | 0.9850 | |
|
| 0.0108 | 0.5821 | 1000 | 0.0802 | 0.8457 | 0.8634 | 0.8544 | 0.9846 | |
|
| 0.0115 | 0.8731 | 1500 | 0.0724 | 0.8486 | 0.8693 | 0.8588 | 0.9850 | |
|
| 0.0087 | 1.1641 | 2000 | 0.0877 | 0.8448 | 0.8680 | 0.8562 | 0.9841 | |
|
| 0.008 | 1.4552 | 2500 | 0.0810 | 0.8576 | 0.8645 | 0.8610 | 0.9850 | |
|
| 0.0084 | 1.7462 | 3000 | 0.0844 | 0.8508 | 0.8661 | 0.8584 | 0.9846 | |
|
| 0.0081 | 2.0373 | 3500 | 0.0859 | 0.8524 | 0.8634 | 0.8579 | 0.9846 | |
|
| 0.0066 | 2.3283 | 4000 | 0.0920 | 0.8386 | 0.8778 | 0.8577 | 0.9845 | |
|
| 0.0065 | 2.6193 | 4500 | 0.0907 | 0.8480 | 0.8624 | 0.8551 | 0.9841 | |
|
| 0.0074 | 2.9104 | 5000 | 0.0873 | 0.8342 | 0.8665 | 0.8500 | 0.9838 | |
|
| 0.0056 | 3.2014 | 5500 | 0.0997 | 0.8490 | 0.8602 | 0.8546 | 0.9842 | |
|
| 0.0056 | 3.4924 | 6000 | 0.0908 | 0.8451 | 0.8696 | 0.8571 | 0.9847 | |
|
| 0.0049 | 3.7835 | 6500 | 0.0991 | 0.8534 | 0.8621 | 0.8577 | 0.9849 | |
|
| 0.0053 | 4.0745 | 7000 | 0.1107 | 0.8444 | 0.8641 | 0.8541 | 0.9844 | |
|
| 0.0041 | 4.3655 | 7500 | 0.0994 | 0.8414 | 0.8658 | 0.8534 | 0.9839 | |
|
| 0.0047 | 4.6566 | 8000 | 0.1016 | 0.8498 | 0.8664 | 0.8580 | 0.9839 | |
|
| 0.0052 | 4.9476 | 8500 | 0.1000 | 0.8369 | 0.8746 | 0.8554 | 0.9844 | |
|
| 0.0039 | 5.2386 | 9000 | 0.1097 | 0.8292 | 0.8777 | 0.8527 | 0.9835 | |
|
| 0.0042 | 5.5297 | 9500 | 0.1000 | 0.8451 | 0.8703 | 0.8575 | 0.9844 | |
|
| 0.0042 | 5.8207 | 10000 | 0.1087 | 0.8468 | 0.8612 | 0.8539 | 0.9841 | |
|
| 0.0042 | 6.1118 | 10500 | 0.1120 | 0.8312 | 0.8753 | 0.8527 | 0.9839 | |
|
| 0.0032 | 6.4028 | 11000 | 0.1144 | 0.8510 | 0.8540 | 0.8525 | 0.9842 | |
|
| 0.0039 | 6.6938 | 11500 | 0.1075 | 0.8330 | 0.8689 | 0.8506 | 0.9837 | |
|
| 0.0038 | 6.9849 | 12000 | 0.1112 | 0.8295 | 0.8677 | 0.8482 | 0.9833 | |
|
| 0.0031 | 7.2759 | 12500 | 0.1084 | 0.8506 | 0.8628 | 0.8567 | 0.9845 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.1.1+cu121 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.19.1 |
|
|
