---
license: mit
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: MiniLM-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. -->

# MiniLM-evidence-types

This model is a fine-tuned version of [microsoft/MiniLM-L12-H384-uncased](https://huggingface.co/microsoft/MiniLM-L12-H384-uncased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.1769
- Macro f1: 0.4136
- Weighted f1: 0.6948
- Accuracy: 0.7017
- Balanced accuracy: 0.3972

## 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: 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 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:--------:|:-----------------:|
| 1.2712        | 1.0   | 250  | 1.1047          | 0.3342   | 0.6582      | 0.7260   | 0.3356            |
| 0.968         | 2.0   | 500  | 0.9558          | 0.3972   | 0.6866      | 0.6948   | 0.4063            |
| 0.8119        | 3.0   | 750  | 1.0086          | 0.3156   | 0.6913      | 0.7002   | 0.3292            |
| 0.6873        | 4.0   | 1000 | 1.0305          | 0.3884   | 0.7035      | 0.7123   | 0.3780            |
| 0.5959        | 5.0   | 1250 | 1.1257          | 0.3922   | 0.6727      | 0.6773   | 0.4151            |
| 0.5078        | 6.0   | 1500 | 1.1642          | 0.3911   | 0.6767      | 0.6773   | 0.4180            |
| 0.4042        | 7.0   | 1750 | 1.2840          | 0.4195   | 0.6891      | 0.6941   | 0.4103            |
| 0.3446        | 8.0   | 2000 | 1.4170          | 0.4208   | 0.6791      | 0.6796   | 0.4240            |
| 0.2973        | 9.0   | 2250 | 1.5195          | 0.4147   | 0.6841      | 0.6849   | 0.4137            |
| 0.2377        | 10.0  | 2500 | 1.6252          | 0.4235   | 0.6950      | 0.7024   | 0.4098            |
| 0.2074        | 11.0  | 2750 | 1.7327          | 0.4139   | 0.6856      | 0.6910   | 0.4046            |
| 0.1849        | 12.0  | 3000 | 1.7941          | 0.4228   | 0.7005      | 0.7070   | 0.4102            |
| 0.146         | 13.0  | 3250 | 1.8656          | 0.4317   | 0.7085      | 0.7199   | 0.4086            |
| 0.137         | 14.0  | 3500 | 2.0057          | 0.4085   | 0.6987      | 0.7040   | 0.4011            |
| 0.1324        | 15.0  | 3750 | 2.0904          | 0.4061   | 0.6822      | 0.6849   | 0.3972            |
| 0.109         | 16.0  | 4000 | 2.0957          | 0.4133   | 0.6908      | 0.6941   | 0.4020            |
| 0.0933        | 17.0  | 4250 | 2.1307          | 0.4011   | 0.6950      | 0.7017   | 0.3876            |
| 0.0925        | 18.0  | 4500 | 2.1606          | 0.4122   | 0.6964      | 0.7040   | 0.3970            |
| 0.0699        | 19.0  | 4750 | 2.1809          | 0.4161   | 0.7004      | 0.7078   | 0.3987            |
| 0.0764        | 20.0  | 5000 | 2.1769          | 0.4136   | 0.6948      | 0.7017   | 0.3972            |


### Framework versions

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