pranay-j/bert-base-uncased-google-boolq

Bert Base Uncased Boolean Question Answer model

This model is a fine-tuned version of bert-base-uncased on the boolq dataset. It achieves the following results on the evaluation set:

• Loss: 0.1993
• Accuracy: 0.7150

Model description

• Model type: Text Classification model
• Language(s) (NLP): English
• License: Apache 2.0

Intended uses & limitations

More information needed

Training and evaluation data

• Dataset

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: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2317	0.9966	147	0.2198	0.6569
0.2	2.0	295	0.2002	0.6960
0.1741	2.9966	442	0.1968	0.7122
0.1469	3.9864	588	0.1993	0.7150

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.2+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1