# Load model directly
from transformers import AutoTokenizer, AutoModelForTokenClassification
tokenizer = AutoTokenizer.from_pretrained("spraxx/bert-base-cased-conll2003-ner")
model = AutoModelForTokenClassification.from_pretrained("spraxx/bert-base-cased-conll2003-ner")Quick Links
bert-base-cased-conll2003-ner
This model is a fine-tuned version of bert-base-cased on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1194
- F1: 0.9116
- Precision: 0.9053
- Recall: 0.9178
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: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall |
|---|---|---|---|---|---|---|
| 0.1659 | 1.0 | 878 | 0.0414 | 0.9342 | 0.9314 | 0.9371 |
| 0.0279 | 2.0 | 1756 | 0.0383 | 0.9480 | 0.9463 | 0.9497 |
| 0.0145 | 3.0 | 2634 | 0.0374 | 0.9518 | 0.9497 | 0.9539 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 2.21.0
- Tokenizers 0.22.2
- Downloads last month
- 46
Model tree for spraxx/bert-base-cased-conll2003-ner
Base model
google-bert/bert-base-casedDataset used to train spraxx/bert-base-cased-conll2003-ner
Evaluation results
- F1 on conll2003validation set self-reported0.912
- Precision on conll2003validation set self-reported0.905
- Recall on conll2003validation set self-reported0.918
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="spraxx/bert-base-cased-conll2003-ner")