bert-base-cased-qa-mash-covid

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

• Loss: 0.4945
• Exact Match: 0.0
• F1: 0.0

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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Exact Match	F1
0.9127	1.0	2820	0.5115	0.0	0.0
0.7236	2.0	5640	0.4775	0.0	0.0
0.5733	3.0	8460	0.4945	0.0	0.0

Framework versions

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