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

Ner

Ner is a Python toolkit for named-entity recognition, providing neural-network architectures and a pretrained CNN model that identifies people, locations and organizations in Russian text.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Ner — 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 it when working with Russian text that can be tokenized, lemmatized and lowercased before inference, or when examining the repository's documented NER training workflow.

Who it is for

  • Developers processing Russian text who need person, location and organization labels.
  • Python users who want to call a named-entity extractor as a module.
  • Researchers or practitioners exploring or training named-entity recognition networks using the repository's documented data format.

Documented uses

  • Tag Russian text with person, location and organization labels from a command line.
  • Retrieve named-entity matches and spans through a Python module.
  • Consult the training notebook for network-training guidance and data formatting.
Before you choose

Requirements & limits

  • Input preprocessing is required: text must be tokenized, lemmatized and lowercased.
  • The repository says it is outdated and directs users to DeepPavlov.
  • The README documents a pretrained CNN model, while its architecture discussion also describes Bi-LSTM+CRF work.
  • Published F1 scores are reported for specific named datasets, not as general performance claims.
  • The repository metadata lists an Apache-2.0 license.

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

Run the pretrained Russian model from the command line

Install the toolkit and use its documented CLI for Russian named-entity recognition.

Before you start

  • Python 3 and the packages listed in the repository's requirements.txt.
  • Text prepared by tokenizing, lemmatizing and lowercasing it.

Steps

  1. Install dependencies with pip3 install -r requirements.txt, or install from the documented GitHub repository URL.

  2. Prepare Russian text by tokenizing, lemmatizing and lowercasing it.

  3. Pass the prepared text to ner.py through standard input, or start ner.py for interactive use.

Expected outcome

The CLI prints tokens with entity tags, including B-LOC and B-ORG in the documented example.

Guide limitations

  • The README says the repository is outdated and points to DeepPavlov.
  • The requirements file does not specify a GPU version of TensorFlow.
Documentation 1 ↗
Source-based practical guide

Use the pretrained extractor from Python

Call the documented Python module to obtain recognized entity spans.

Before you start

  • Python 3 and the toolkit's required packages.
  • Text prepared by tokenizing, lemmatizing and lowercasing it.

Steps

  1. Install the toolkit dependencies using the repository's documented installation instructions.

  2. Prepare the input text as required by the README.

  3. Import ner, create an ner.Extractor instance and call it with the prepared text.

Expected outcome

The example returns entity matches with token spans and entity types such as LOC and ORG.

Guide limitations

  • The repository describes itself as outdated and points to DeepPavlov.
  • The README's example demonstrates Russian text and the pretrained Russian model.
Documentation 1 ↗
Evidence and transparency

Sources & review

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