bert-banking-intent / README.md
hf-tuner's picture
Update README.md
948fbc3 verified
metadata
library_name: transformers
license: apache-2.0
base_model: google-bert/bert-base-uncased
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: bert-banking-intent
    results: []
datasets:
  - hf-tuner/banking-intent
language:
  - en
pipeline_tag: text-classification

bert-banking-intent

This model is a fine-tuned version of google-bert/bert-base-uncased on hf-tuner/banking-intent dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0079
  • Accuracy: 0.9993

How to Get Started with the Model


from transformers import pipeline

classifier = pipeline("text-classification", model = "hf-tuner/bert-banking-intent")
classifier("Please help me get a new card, I reside in the United States.")
## [{'label': 'country_support', 'score': 0.997}]

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
1.9901 1.0 626 1.5437 0.8104
0.8228 2.0 1252 0.5328 0.9335
0.3901 3.0 1878 0.2214 0.9678
0.1889 4.0 2504 0.1041 0.9830
0.0973 5.0 3130 0.0518 0.9920
0.0733 6.0 3756 0.0322 0.9944
0.0405 7.0 4382 0.0167 0.9976
0.0214 8.0 5008 0.0114 0.9988
0.0175 9.0 5634 0.0091 0.9993
0.0138 10.0 6260 0.0079 0.9993

Framework versions

  • Transformers 4.57.1
  • Pytorch 2.8.0+cu126
  • Datasets 4.0.0
  • Tokenizers 0.22.1