Edit model card

bert-small-finetuned-xglue-ner-longer20

This model is a fine-tuned version of muhtasham/bert-small-finetuned-xglue-ner-longer10 on the wnut_17 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5839
  • Precision: 0.5783
  • Recall: 0.4330
  • F1: 0.4952
  • Accuracy: 0.9261

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 1.0 425 0.4882 0.5 0.4510 0.4742 0.9228
0.013 2.0 850 0.5039 0.5130 0.4246 0.4647 0.9242
0.0124 3.0 1275 0.5331 0.5506 0.4426 0.4907 0.9256
0.0159 4.0 1700 0.5403 0.5488 0.4234 0.4781 0.9249
0.0132 5.0 2125 0.5459 0.5714 0.4211 0.4848 0.9255
0.0117 6.0 2550 0.5522 0.5637 0.4342 0.4905 0.9253
0.0117 7.0 2975 0.5712 0.5778 0.4354 0.4966 0.9257
0.007 8.0 3400 0.5860 0.5828 0.4378 0.5 0.9259
0.0066 9.0 3825 0.5745 0.5703 0.4462 0.5007 0.9259
0.0049 10.0 4250 0.5839 0.5783 0.4330 0.4952 0.9261

Framework versions

  • Transformers 4.21.1
  • Pytorch 1.12.1+cu113
  • Datasets 2.4.0
  • Tokenizers 0.12.1
Downloads last month
5
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.

Dataset used to train muhtasham/bert-small-finetuned-xglue-ner-longer20

Evaluation results