NuNerZero_onnx Open-source Model - Quickly and Accurately Identify Entities, Ultra-efficient with Zero-shot Usability

Nunerzero Onnx

Developed by deepanwa

The ONNX version of NuNerZero, a zero-shot named entity recognition model optimized for fast inference using ONNX Runtime.

Sequence Labeling Open Source License:MIT #Zero-shot NER #ONNX acceleration #Medical anonymization

Downloads 174

Release Time : 3/22/2025

Model Overview

This is a zero-shot named entity recognition (NER) model, converted to ONNX format to provide efficient inference performance, suitable for production environments.

Model Features

Efficient Inference

Optimized through ONNX format, significantly improving inference speed

Zero-shot Capability

Can recognize new entity types without domain-specific training

Production-ready

The model is optimized and ready for direct use in production environments

Model Capabilities

Zero-shot named entity recognition

Multi-category entity recognition

Text information extraction

Use Cases

Data anonymization

Personal information anonymization

Identify and anonymize sensitive personal information in text

Can effectively recognize sensitive information such as names, phone numbers, dates, etc.

Information extraction

Medical record analysis

Extract entities such as patient information and diagnosis results from medical texts

🚀 ONNX Model for NuNerZero

This repository stores the ONNX version of NuNerZero, a zero - shot named entity recognition (NER) model. It is optimized for fast inference using ONNX Runtime. The conversion aims to offer efficient and production - ready performance while retaining the original capabilities of the NuNerZero model.

This model is a part of Zink. Zink is a zero - shot anonymizer and currently uses the ONNX NuNERZero model for anonymization.

🚀 Quick Start

Here’s a quick example on how to load and use the ONNX model with GLiNER:

from gliner import GLiNER
import time

# Load the ONNX model and tokenizer
model_name="deepanwa/NuNerZero_onnx"
model = GLiNER.from_pretrained(model_name,load_onnx_model=True, load_tokenizer=True)

text = "Dr. Michael, a cardiologist from Canada, was born on 07/04/1970. John Doe dialled his mother at 992-234-3456 and then went out for a walk."
labels = ("person", "profession", "location", "date", "phone number", "relationship", "medical condition", "age")

start = time.time()
result = model.predict_entities(text, labels)
end = time.time()

print("Predicted entities:", result)
print("Time taken:", end - start)

✨ Features

Zero - shot named entity recognition.
Optimized for fast inference using ONNX Runtime.
Part of the Zink zero - shot anonymizer.

📦 Installation

Requirements

Python 3.7 or higher
GLiNER – the package that provides the interface to load and run the model. The ONNX version was created on "gliner==0.2.3".

📚 Documentation

Repository Contents

Property	Details
model.onnx	The primary ONNX model file used for inference.
gliner_config.json	Configuration settings for the model.
added_tokens.json	Additional tokens required by the tokenizer.
special_tokens_map.json	Mapping for special tokens.
tokenizer.json and tokenizer_config.json	Tokenizer vocabulary and configuration files.
spm.model	SentencePiece model file used by the tokenizer.

🔧 Technical Details

The model is a zero - shot NER model.
It is converted to the ONNX format to improve inference speed.
It is used as a component in the Zink zero - shot anonymizer.

📄 License

This project is licensed under the MIT license.

📄 Citation

10.57967/hf/4902

💬 Contributing

Contributions, suggestions, or bug reports are welcome. Please open an issue or submit a pull request if you have improvements.

⚠️ Important Note

The model file is approximately 1.85 GB, so please ensure you have sufficient bandwidth and disk space when downloading.

💡 Usage Tip

Leveraging the ONNX format can significantly accelerate inference compared to the original PyTorch implementation.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご