distillbert-fine-tune-ner-task
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2001
- Precision: 0.7185
- Recall: 0.9406
- F1: 0.8147
- Accuracy: 0.9768
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: 1
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.3931 | 0.14 | 500 | 0.2367 | 0.6379 | 0.9153 | 0.7518 | 0.9693 |
0.1874 | 0.28 | 1000 | 0.2640 | 0.6635 | 0.9227 | 0.7719 | 0.9720 |
0.1943 | 0.42 | 1500 | 0.2230 | 0.6974 | 0.9265 | 0.7958 | 0.9756 |
0.165 | 0.56 | 2000 | 0.2426 | 0.7017 | 0.9308 | 0.8002 | 0.9754 |
0.1461 | 0.71 | 2500 | 0.2327 | 0.7087 | 0.9333 | 0.8056 | 0.9763 |
0.1634 | 0.85 | 3000 | 0.2139 | 0.7121 | 0.9372 | 0.8093 | 0.9761 |
0.1533 | 0.99 | 3500 | 0.2001 | 0.7185 | 0.9406 | 0.8147 | 0.9768 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu118
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 4