|
--- |
|
license: mit |
|
base_model: roberta-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- recall |
|
- precision |
|
- f1 |
|
model-index: |
|
- name: roberta-base-suicide-prediction-phr-v2 |
|
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. --> |
|
|
|
# PHR_Suicide_Prediction_Roberta_Cleaned_Light |
|
|
|
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: 0.0553 |
|
- Accuracy: 0.9869 |
|
- Recall: 0.9846 |
|
- Precision: 0.9904 |
|
- F1: 0.9875 |
|
|
|
## 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: 16 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.06 |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Recall | Precision | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:| |
|
| 0.1928 | 0.05 | 500 | 0.2289 | 0.9340 | 0.9062 | 0.9660 | 0.9352 | |
|
| 0.0833 | 0.1 | 1000 | 0.1120 | 0.9752 | 0.9637 | 0.9888 | 0.9761 | |
|
| 0.0366 | 0.16 | 1500 | 0.1165 | 0.9753 | 0.9613 | 0.9915 | 0.9762 | |
|
| 0.071 | 0.21 | 2000 | 0.0973 | 0.9709 | 0.9502 | 0.9940 | 0.9716 | |
|
| 0.0465 | 0.26 | 2500 | 0.0680 | 0.9829 | 0.9979 | 0.9703 | 0.9839 | |
|
| 0.0387 | 0.31 | 3000 | 0.1583 | 0.9705 | 0.9490 | 0.9945 | 0.9712 | |
|
| 0.1061 | 0.37 | 3500 | 0.0685 | 0.9848 | 0.9802 | 0.9907 | 0.9854 | |
|
| 0.0593 | 0.42 | 4000 | 0.0550 | 0.9872 | 0.9947 | 0.9813 | 0.9879 | |
|
| 0.0382 | 0.47 | 4500 | 0.0551 | 0.9871 | 0.9912 | 0.9842 | 0.9877 | |
|
| 0.0831 | 0.52 | 5000 | 0.0502 | 0.9840 | 0.9768 | 0.9927 | 0.9847 | |
|
| 0.0376 | 0.58 | 5500 | 0.0654 | 0.9865 | 0.9852 | 0.9889 | 0.9871 | |
|
| 0.0634 | 0.63 | 6000 | 0.0422 | 0.9877 | 0.9897 | 0.9870 | 0.9883 | |
|
| 0.0235 | 0.68 | 6500 | 0.0553 | 0.9869 | 0.9846 | 0.9904 | 0.9875 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.0 |
|
|
