metadata
license: apache-2.0
language: en
tags:
- named-entity-recognition
- distilbert
- transformers
datasets:
- conll2003
metrics:
- f1
- accuracy
- precision
- recall
model_type: distilbert
Fine-tuned DistilBERT for Named Entity Recognition (NER)
Model Description
This model is a fine-tuned version of DistilBERT for Named Entity Recognition (NER) tasks. It was trained on the CoNLL-2003 dataset, designed to identify entities such as persons, organizations, locations, and miscellaneous entities within English text.
- Model Architecture: DistilBERT (pre-trained transformer-based model)
- Task: Named Entity Recognition (NER)
- Entity Types: PER (Person), ORG (Organization), LOC (Location), MISC (Miscellaneous)
Training Details
- Dataset: CoNLL-2003 (standard dataset for NER tasks)
- Training Data Size: 14,000 samples for training, 3,250 samples for evaluation
- Epochs: 3
- Batch Size: 16 (training), 64 (evaluation)
- Learning Rate: 2e-5
- Optimizer: AdamW with weight decay
Evaluation Metrics
The model was evaluated using the following metrics:
- F1 Score: 0.928661
- Accuracy: 0.983252
- Precision: 0.918794
- Recall: 0.938741
Example Usage
Here’s how to use this NER model with the Hugging Face Transformers library:
from transformers import pipeline
# Load the model from the Hugging Face Hub
ner_pipeline = pipeline("ner", model="Beehzod/smart-finetuned-ner")
# Example predictions
text = "Hugging Face Inc. is based in New York City, and its CEO is Clement Delangue."
results = ner_pipeline(text)
for entity in results:
print(f"Entity: {entity['word']}, Label: {entity['entity']}, Score: {entity['score']:.4f}")