---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: BERT-evidence-types
  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. -->

# BERT-evidence-types

This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.9735
- Macro f1: 0.3791
- Weighted f1: 0.6925
- Accuracy: 0.7070
- Balanced accuracy: 0.3625

## 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: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Macro f1 | Weighted f1 | Accuracy | Balanced accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:--------:|:-----------------:|
| 1.098         | 1.0   | 250  | 1.0176          | 0.2666   | 0.6861      | 0.7070   | 0.2775            |
| 0.7656        | 2.0   | 500  | 1.0072          | 0.4124   | 0.7126      | 0.7215   | 0.3876            |
| 0.5045        | 3.0   | 750  | 1.1791          | 0.3759   | 0.6843      | 0.6910   | 0.3799            |
| 0.2874        | 4.0   | 1000 | 1.4338          | 0.3738   | 0.6888      | 0.6986   | 0.3705            |
| 0.1599        | 5.0   | 1250 | 1.8058          | 0.3839   | 0.6947      | 0.7070   | 0.3682            |
| 0.0991        | 6.0   | 1500 | 2.0263          | 0.3777   | 0.6793      | 0.6903   | 0.3627            |
| 0.0529        | 7.0   | 1750 | 2.2380          | 0.4046   | 0.6932      | 0.7047   | 0.3877            |
| 0.0311        | 8.0   | 2000 | 2.4153          | 0.4185   | 0.6999      | 0.7131   | 0.3899            |
| 0.0129        | 9.0   | 2250 | 2.7230          | 0.3702   | 0.6852      | 0.7123   | 0.3331            |
| 0.0102        | 10.0  | 2500 | 2.6453          | 0.4115   | 0.6934      | 0.7070   | 0.3880            |
| 0.0141        | 11.0  | 2750 | 2.7078          | 0.4054   | 0.6859      | 0.6979   | 0.3863            |
| 0.0088        | 12.0  | 3000 | 2.7182          | 0.3724   | 0.6904      | 0.7062   | 0.3559            |
| 0.0061        | 13.0  | 3250 | 2.7814          | 0.4091   | 0.6917      | 0.7055   | 0.3839            |
| 0.0069        | 14.0  | 3500 | 2.8035          | 0.3836   | 0.6986      | 0.7108   | 0.3688            |
| 0.0067        | 15.0  | 3750 | 2.9326          | 0.4119   | 0.6952      | 0.7139   | 0.3793            |
| 0.0049        | 16.0  | 4000 | 2.9338          | 0.4133   | 0.6885      | 0.7040   | 0.3794            |
| 0.0065        | 17.0  | 4250 | 2.9380          | 0.3820   | 0.6964      | 0.7100   | 0.3650            |
| 0.0045        | 18.0  | 4500 | 2.9439          | 0.3802   | 0.6925      | 0.7055   | 0.3646            |
| 0.0044        | 19.0  | 4750 | 2.9731          | 0.3796   | 0.6932      | 0.7078   | 0.3626            |
| 0.0056        | 20.0  | 5000 | 2.9735          | 0.3791   | 0.6925      | 0.7070   | 0.3625            |


### Framework versions

- Transformers 4.19.2
- Pytorch 1.11.0+cu113
- Datasets 2.2.2
- Tokenizers 0.12.1