roberta

This model is a fine-tuned version of Jean-Baptiste/roberta-large-ner-english on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3908
  • Precision: 0.5990
  • Recall: 0.5581
  • F1: 0.5778
  • Accuracy: 0.9470

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: 32
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.99) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.0 151 0.2078 0.1899 0.2388 0.2115 0.9246
No log 2.0 302 0.1499 0.4322 0.5535 0.4854 0.9393
No log 3.0 453 0.1916 0.5204 0.4946 0.5072 0.9418
0.1542 4.0 604 0.1671 0.4615 0.5109 0.4849 0.9426
0.1542 5.0 755 0.1940 0.4841 0.4829 0.4835 0.9439
0.1542 6.0 906 0.2462 0.5066 0.5651 0.5343 0.9428
0.0616 7.0 1057 0.2106 0.5041 0.5271 0.5153 0.9437
0.0616 8.0 1208 0.2621 0.5620 0.5202 0.5403 0.9474
0.0616 9.0 1359 0.2903 0.5242 0.5550 0.5392 0.9440
0.0326 10.0 1510 0.3083 0.5883 0.5628 0.5753 0.9483
0.0326 11.0 1661 0.3125 0.5451 0.5853 0.5645 0.9444
0.0326 12.0 1812 0.3616 0.5503 0.5388 0.5445 0.9427
0.0326 13.0 1963 0.3398 0.5978 0.5023 0.5459 0.9447
0.0155 14.0 2114 0.2942 0.5701 0.5550 0.5625 0.9467
0.0155 15.0 2265 0.3723 0.5771 0.5597 0.5683 0.9462
0.0155 16.0 2416 0.3651 0.5751 0.5760 0.5755 0.9439
0.0062 17.0 2567 0.3674 0.5667 0.5891 0.5777 0.9455
0.0062 18.0 2718 0.3866 0.5897 0.5403 0.5639 0.9463
0.0062 19.0 2869 0.3908 0.5990 0.5581 0.5778 0.9470
0.0033 20.0 3020 0.4036 0.5914 0.5620 0.5763 0.9467

Framework versions

  • Transformers 4.31.0
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.2
  • Tokenizers 0.13.3
Downloads last month
29
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 autosyrup/roberta

Finetuned
(1)
this model