---
license: mit
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: PubMedBERT-LitCovid-1.4
  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. -->

# PubMedBERT-LitCovid-1.4

This model is a fine-tuned version of [microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract](https://huggingface.co/microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5628
- Hamming loss: 0.0745
- F1 micro: 0.6343
- F1 macro: 0.4913
- F1 weighted: 0.7105
- F1 samples: 0.6391
- Precision micro: 0.4918
- Precision macro: 0.3747
- Precision weighted: 0.6260
- Precision samples: 0.5363
- Recall micro: 0.8930
- Recall macro: 0.8406
- Recall weighted: 0.8930
- Recall samples: 0.9098
- Roc Auc: 0.9106
- Accuracy: 0.0952

## 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
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Hamming loss | F1 micro | F1 macro | F1 weighted | F1 samples | Precision micro | Precision macro | Precision weighted | Precision samples | Recall micro | Recall macro | Recall weighted | Recall samples | Roc Auc | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:------------:|:--------:|:--------:|:-----------:|:----------:|:---------------:|:---------------:|:------------------:|:-----------------:|:------------:|:------------:|:---------------:|:--------------:|:-------:|:--------:|
| 0.6486        | 1.0   | 1151 | 0.6207          | 0.1099       | 0.5362   | 0.4107   | 0.6522      | 0.5433     | 0.3858          | 0.3021          | 0.5651             | 0.4237            | 0.8791       | 0.8500       | 0.8791          | 0.8964         | 0.8850  | 0.0234   |
| 0.5189        | 2.0   | 2303 | 0.5572          | 0.0981       | 0.5696   | 0.4299   | 0.6739      | 0.5815     | 0.4170          | 0.3178          | 0.5825             | 0.4655            | 0.8984       | 0.8672       | 0.8984          | 0.9143         | 0.9002  | 0.0501   |
| 0.4426        | 3.0   | 3454 | 0.5516          | 0.0853       | 0.6029   | 0.4632   | 0.6947      | 0.6086     | 0.4545          | 0.3493          | 0.6085             | 0.4966            | 0.8951       | 0.8538       | 0.8951          | 0.9116         | 0.9057  | 0.0650   |
| 0.3771        | 4.0   | 4606 | 0.5647          | 0.0735       | 0.6371   | 0.4944   | 0.7110      | 0.6402     | 0.4955          | 0.3779          | 0.6258             | 0.5377            | 0.8920       | 0.8363       | 0.8920          | 0.9087         | 0.9106  | 0.0924   |
| 0.3467        | 5.0   | 5755 | 0.5628          | 0.0745       | 0.6343   | 0.4913   | 0.7105      | 0.6391     | 0.4918          | 0.3747          | 0.6260             | 0.5363            | 0.8930       | 0.8406       | 0.8930          | 0.9098         | 0.9106  | 0.0952   |


### Framework versions

- Transformers 4.28.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3