clamsproject/bert-base-cased-ner-rfb

This model is a fine-tuned version of bert-base-cased on a private dataset.

It achieves the following results on the evaluation set:

• eval_loss: 1.2720
• eval_FILL_precision: 0.7627
• eval_FILL_recall: 0.7759
• eval_FILL_f1: 0.7692
• eval_FILL_number: 58
• eval_ROLE_precision: 0.8125
• eval_ROLE_recall: 0.8125
• eval_ROLE_f1: 0.8125
• eval_ROLE_number: 48
• eval_overall_precision: 0.7850
• eval_overall_recall: 0.7925
• eval_overall_f1: 0.7887
• eval_overall_accuracy: 0.8289
• eval_runtime: 1.3592
• eval_samples_per_second: 44.144
• eval_steps_per_second: 5.886
• step: 0

It achieves the following results on the test set:

• test_FILL_f1: 0.8039
• test_FILL_number: 46,
• test_FILL_precision: 0.7321
• test_FILL_recall: 0.8913
• test_ROLE_f1: 0.8182
• test_ROLE_number: 42,
• test_ROLE_precision: 0.7826
• test_ROLE_recall: 0.8571
• test_loss: 0.9132
• test_overall_accuracy: 0.8791
• test_overall_f1: 0.8105
• test_overall_precision: 0.7549
• test_overall_recall: 0.875
• test_runtime: 0.9583
• test_samples_per_second: 63.652
• test_steps_per_second: 8.348

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 600

Framework versions

• Transformers 4.40.1
• Pytorch 2.2.1+cu121
• Datasets 2.14.6
• Tokenizers 0.19.1