distilbert-finetuned-headings

This model is a fine-tuned version of distilbert-base-cased on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1669
  • F1 Positive: 0.9112
  • F1 Negative: 0.9854
  • F1: 0.9749
  • Roc Auc: 0.9457
  • Accuracy: 0.9749

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: 1e-05
  • train_batch_size: 8
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 12

Training results

Training Loss Epoch Step Validation Loss F1 Positive F1 Negative F1 Roc Auc Accuracy
0.1869 1.0 1785 0.1452 0.8621 0.9793 0.9640 0.8922 0.9640
0.1306 2.0 3570 0.1190 0.8738 0.9807 0.9665 0.9031 0.9665
0.1182 3.0 5355 0.1460 0.8831 0.9818 0.9685 0.9137 0.9685
0.0841 4.0 7140 0.1431 0.8990 0.9844 0.9730 0.9201 0.9730
0.061 5.0 8925 0.1540 0.9066 0.9846 0.9736 0.9431 0.9736
0.0381 6.0 10710 0.1630 0.9070 0.9851 0.9743 0.9359 0.9743
0.0268 7.0 12495 0.1669 0.9112 0.9854 0.9749 0.9457 0.9749
0.024 8.0 14280 0.2216 0.8964 0.9827 0.9704 0.9412 0.9704
0.0182 9.0 16065 0.2294 0.9032 0.9843 0.9730 0.9371 0.9730
0.0176 10.0 17850 0.2239 0.9057 0.9847 0.9736 0.9393 0.9736
0.0197 11.0 19635 0.2441 0.8966 0.9832 0.9710 0.9340 0.9710
0.0128 12.0 21420 0.2541 0.8899 0.9820 0.9691 0.9310 0.9691

Framework versions

  • Transformers 4.34.1
  • Pytorch 2.1.0+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1
Downloads last month
21
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for OrbitalWitness/distilbert-finetuned-headings-v2

Finetuned
(224)
this model