<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
roberta-base-ner
This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0693
- Overall Precision: 0.7067
- Overall Recall: 0.6726
- Overall F1: 0.6892
- Org Precision: 0.7504
- Org Recall: 0.6531
- Org F1: 0.6984
- Per Precision: 0.7525
- Per Recall: 0.8736
- Per F1: 0.8085
- Loc Precision: 0.4769
- Loc Recall: 0.6263
- Loc F1: 0.5415
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Overall Precision | Overall Recall | Overall F1 | Org Precision | Org Recall | Org F1 | Per Precision | Per Recall | Per F1 | Loc Precision | Loc Recall | Loc F1 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 53 | 0.0935 | 0.5785 | 0.4382 | 0.4987 | 0.5785 | 0.5611 | 0.5697 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| No log | 2.0 | 106 | 0.0693 | 0.7067 | 0.6726 | 0.6892 | 0.7504 | 0.6531 | 0.6984 | 0.7525 | 0.8736 | 0.8085 | 0.4769 | 0.6263 | 0.5415 |
| No log | 3.0 | 159 | 0.0994 | 0.7832 | 0.6466 | 0.7084 | 0.8154 | 0.5928 | 0.6865 | 0.9444 | 0.9770 | 0.9605 | 0.5504 | 0.7172 | 0.6228 |
| No log | 4.0 | 212 | 0.0761 | 0.7872 | 0.6973 | 0.7395 | 0.7855 | 0.6516 | 0.7123 | 0.9444 | 0.9770 | 0.9605 | 0.6696 | 0.7576 | 0.7109 |
| No log | 5.0 | 265 | 0.0785 | 0.7932 | 0.6914 | 0.7388 | 0.7922 | 0.6440 | 0.7105 | 0.9444 | 0.9770 | 0.9605 | 0.6757 | 0.7576 | 0.7143 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
