Edit model card

roberta-finetuned-ner-without-data-sort

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

  • Loss: 0.0420
  • Precision: 0.9914
  • Recall: 0.9909
  • F1: 0.9912
  • Accuracy: 0.9920

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: 2e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.0 213 0.1879 0.9378 0.9414 0.9396 0.9493
No log 2.0 426 0.1038 0.9725 0.9750 0.9737 0.9751
0.4424 3.0 639 0.0701 0.9861 0.9851 0.9856 0.9863
0.4424 4.0 852 0.0637 0.9882 0.9880 0.9881 0.9880
0.0675 5.0 1065 0.0546 0.9851 0.9878 0.9865 0.9879
0.0675 6.0 1278 0.0480 0.9894 0.9904 0.9899 0.9901
0.0675 7.0 1491 0.0473 0.9919 0.9904 0.9912 0.9911
0.0426 8.0 1704 0.0441 0.9921 0.9916 0.9919 0.9921
0.0426 9.0 1917 0.0426 0.9921 0.9916 0.9919 0.9922
0.033 10.0 2130 0.0420 0.9914 0.9909 0.9912 0.9920

Framework versions

  • Transformers 4.17.0
  • Pytorch 1.10.0+cu111
  • Datasets 2.0.0
  • Tokenizers 0.11.6
Downloads last month
13