xlm-roberta-base-ner-hrl-ner-finetuning
This model is a fine-tuned version of Davlan/xlm-roberta-base-ner-hrl on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1135
- Precision: 0.9290
- Recall: 0.9367
- F1: 0.9328
- Accuracy: 0.9801
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: 32
- 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 |
|---|---|---|---|---|---|---|---|
| 0.1534 | 1.0 | 477 | 0.0870 | 0.9001 | 0.9124 | 0.9062 | 0.9740 |
| 0.077 | 2.0 | 954 | 0.0764 | 0.9187 | 0.9321 | 0.9253 | 0.9789 |
| 0.0529 | 3.0 | 1431 | 0.0845 | 0.9178 | 0.9313 | 0.9245 | 0.9791 |
| 0.0377 | 4.0 | 1908 | 0.0805 | 0.9200 | 0.9310 | 0.9255 | 0.9795 |
| 0.0292 | 5.0 | 2385 | 0.0918 | 0.9278 | 0.9346 | 0.9312 | 0.9795 |
| 0.0204 | 6.0 | 2862 | 0.1016 | 0.9222 | 0.9323 | 0.9273 | 0.9790 |
| 0.0167 | 7.0 | 3339 | 0.1066 | 0.9271 | 0.9327 | 0.9299 | 0.9790 |
| 0.0134 | 8.0 | 3816 | 0.1088 | 0.9253 | 0.9358 | 0.9305 | 0.9797 |
| 0.0101 | 9.0 | 4293 | 0.1134 | 0.9289 | 0.9357 | 0.9323 | 0.9798 |
| 0.0079 | 10.0 | 4770 | 0.1135 | 0.9290 | 0.9367 | 0.9328 | 0.9801 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 11
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.