Model description
This model is a fine-tuned version of roberta-base for the Named Entity Recognition (NER) task using the CoNLL-2003 dataset. It can identify four types of entities: Persons (PER), Organizations (ORG), Locations (LOC), and Miscellaneous (MISC).
Training procedure
- Hardware: NVIDIA V100 GPU
- Optimizer: AdamW
- Learning Rate: 2e-5
- Batch Size: 16
- Weight Decay: 0.01
- Epochs: 5
- Mixed Precision Training: FP16 enabled
Evaluation Results
| Metric) | Value |
|---|---|
| F1 Score | 95.99% |
| Precision | 95.61% |
| Recall | 96.38% |
| Accuracy | 99.29% |
| Eval Loss | 0.0464 |
How to use
from transformers import pipeline
model_id = "learnrr/roberta-NER-conll2003"
text = "Apple is looking at buying U.K. startup for $1 billion"
results = nlp(text)
for entity in results:
print(f"entity: {entity['word']} | class: {entity['entity_group']} | confidence: {entity['score']:.4f}")