Eurosmart's picture
model tuned
b78ab96 verified
metadata
license: mit
base_model: FacebookAI/roberta-base
tags:
  - question-answering
  - nlp
  - generated_from_trainer
model-index:
  - name: distilbert-qa-mash-covid
    results: []

distilbert-qa-mash-covid

This model is a fine-tuned version of FacebookAI/roberta-base on the mashqa_covid_dataset dataset. It achieves the following results on the evaluation set:

  • Loss: 1.0929

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: 5

Training results

Training Loss Epoch Step Validation Loss
1.7112 1.0 657 1.0613
1.0835 2.0 1314 1.0024
0.9518 3.0 1971 1.0094
0.6736 4.0 2628 1.0287
0.5997 5.0 3285 1.0929

Framework versions

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