---
license: apache-2.0
base_model: google/bigbird-roberta-base
tags:
- question-answering
- nlp
- generated_from_trainer
model-index:
- name: bert-qa-mash-covid
  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-qa-mash-covid

This model is a fine-tuned version of [google/bigbird-roberta-base](https://huggingface.co./google/bigbird-roberta-base) on the mashqa_covid_dataset dataset.
It achieves the following results on the evaluation set:
- Loss: 4.1822

## 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: 50

### Training results

| Training Loss | Epoch | Step  | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 1.5553        | 1.0   | 657   | 1.0374          |
| 1.0747        | 2.0   | 1314  | 0.9805          |
| 0.8656        | 3.0   | 1971  | 1.0557          |
| 0.5949        | 4.0   | 2628  | 1.1109          |
| 0.4757        | 5.0   | 3285  | 1.1982          |
| 0.3937        | 6.0   | 3942  | 1.3565          |
| 0.2476        | 7.0   | 4599  | 1.4817          |
| 0.2064        | 8.0   | 5256  | 1.5591          |
| 0.191         | 9.0   | 5913  | 2.0118          |
| 0.1382        | 10.0  | 6570  | 2.1637          |
| 0.1161        | 11.0  | 7227  | 2.1896          |
| 0.1066        | 12.0  | 7884  | 2.3879          |
| 0.1017        | 13.0  | 8541  | 2.6023          |
| 0.0901        | 14.0  | 9198  | 2.8120          |
| 0.0859        | 15.0  | 9855  | 2.8291          |
| 0.0666        | 16.0  | 10512 | 2.8947          |
| 0.0598        | 17.0  | 11169 | 3.1198          |
| 0.0459        | 18.0  | 11826 | 3.0020          |
| 0.0533        | 19.0  | 12483 | 3.1257          |
| 0.0546        | 20.0  | 13140 | 3.1757          |
| 0.0487        | 21.0  | 13797 | 3.4253          |
| 0.0424        | 22.0  | 14454 | 3.2998          |
| 0.0465        | 23.0  | 15111 | 3.4184          |
| 0.0298        | 24.0  | 15768 | 3.6026          |
| 0.038         | 25.0  | 16425 | 3.7422          |
| 0.032         | 26.0  | 17082 | 3.6208          |
| 0.0247        | 27.0  | 17739 | 3.6113          |
| 0.0283        | 28.0  | 18396 | 3.8404          |
| 0.0237        | 29.0  | 19053 | 3.6326          |
| 0.017         | 30.0  | 19710 | 3.8710          |
| 0.0195        | 31.0  | 20367 | 3.8173          |
| 0.0205        | 32.0  | 21024 | 3.7101          |
| 0.0187        | 33.0  | 21681 | 3.8000          |
| 0.0236        | 34.0  | 22338 | 3.7253          |
| 0.0251        | 35.0  | 22995 | 3.7891          |
| 0.0399        | 36.0  | 23652 | 3.8655          |
| 0.0355        | 37.0  | 24309 | 3.7444          |
| 0.0214        | 38.0  | 24966 | 3.9477          |
| 0.0132        | 39.0  | 25623 | 4.1278          |
| 0.019         | 40.0  | 26280 | 4.1814          |
| 0.0227        | 41.0  | 26937 | 3.9744          |
| 0.0161        | 42.0  | 27594 | 4.1047          |
| 0.0165        | 43.0  | 28251 | 4.2342          |
| 0.0162        | 44.0  | 28908 | 4.1003          |
| 0.013         | 45.0  | 29565 | 4.2922          |
| 0.0158        | 46.0  | 30222 | 4.1349          |
| 0.0118        | 47.0  | 30879 | 4.1534          |
| 0.0155        | 48.0  | 31536 | 4.1656          |
| 0.0167        | 49.0  | 32193 | 4.1763          |
| 0.0205        | 50.0  | 32850 | 4.1822          |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2