OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Nerpy

Nerpy is an open-source Python library for named-entity recognition, implementing BERT-based BertSoftmax, BertSpan and BertCrf approaches with PyTorch. Its documented interface predicts labeled entities from English and Chinese text, and the repository includes training and prediction examples.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Nerpy — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Nerpy when named-entity recognition needs to be a callable part of a Python text-processing workflow, especially when the supplied English or Chinese model examples suit the task.

Who it is for

  • Python developers adding entity extraction to an NLP application.
  • Teams building text-processing pipelines that need labeled people, places, organizations, times or other entities.
  • Practitioners comparing or using the repository's BertSoftmax and BertSpan implementations.

Documented uses

  • Extract entity spans and labels from sentences through a Python prediction call.
  • Choose between documented BertSoftmax and BertSpan model examples for supported recognition workflows.
  • Train or predict with the repository's documented model implementations.
Before you choose

Requirements & limits

The repository is public and identifies its license as Apache-2.0. The README specifies Python 3.8 or later and installation of PyTorch.

  • The documented setup requires Python 3.8+ and PyTorch.
  • The examples load model identifiers from Hugging Face and describe automatic local caching of downloaded models.
  • Benchmark results are reported for named datasets and configurations; the documented GPU throughput environment is a Tesla V100 with 32 GB of memory.
  • The README includes English and Chinese examples, but task fit depends on the model and labels selected.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Nerpy and recognize entities in text

Set up the package and call its documented prediction interface on English or Chinese sentences.

Before you start

  • Python 3.8 or later
  • PyTorch
  • A Hugging Face model identifier; README examples include shibing624/bert4ner-base-uncased and shibing624/bert4ner-base-chinese

Steps

  1. Install PyTorch using pip or conda.

  2. Install Nerpy with pip install -U nerpy.

  3. Import NERModel and initialize it with the documented model type and model identifier.

  4. Call predict with a list of sentences; the examples show split_on_space=True for English and split_on_space=False for Chinese.

Expected outcome

The prediction call returns predictions, raw outputs and entity tuples containing recognized text and labels.

Guide limitations

  • The README describes model weights downloading automatically to a local Hugging Face cache.
  • The guide follows the repository's examples and does not establish performance or compatibility for other environments or models.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
shibing624/nerpy · Official homepage title · retained in its source language
How we review sourcesSuggest a correction