OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Kashgari

Kashgari is a Python NLP transfer-learning framework built on tf.keras for developing text classification and text-labeling models. It supports tasks including named entity recognition, part-of-speech tagging and sequence labeling, and provides Word2Vec, BERT and GPT-2 language embeddings.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Kashgari — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Kashgari when developing Python-based NLP classification or labeling models with a documented tf.keras workflow and embedding options, or when following its tutorials to learn those tasks.

Who it is for

  • NLP developers building classification or text-labeling models.
  • Researchers experimenting with NLP models and approaches.
  • Learners following tutorials to build NLP projects.

Documented uses

  • Train a text classifier for labeled text data.
  • Build a named entity recognition model.
  • Create a part-of-speech tagging or other sequence-labeling model.
  • Experiment with pretrained language embeddings in an NLP project.
  • Prepare a model for TensorFlow Serving by exporting it in SavedModel format.
Before you choose

Requirements & limits

  • Kashgari is a software framework in a public GitHub repository, rather than a hosted model-analysis service.
  • The installation path depends on the selected Kashgari generation and backend; TensorFlow use also requires TensorFlow Addons.
  • The repository documents SavedModel export for TensorFlow Serving, which can inform a development-to-serving workflow.
  • The repository lists Apache-2.0 as its 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

Prepare a Kashgari development environment

Choose a documented Kashgari installation path that matches the TensorFlow or Keras backend.

Before you start

  • Python 3.6 or later, as stated in the repository README.
  • A supported backend and matching Kashgari version; TensorFlow users also need TensorFlow Addons.

Steps

  1. Choose the backend and Kashgari version row in the README installation table: TensorFlow 2.2+ with Kashgari 2.0.2 or later, TensorFlow 1.14+ with Kashgari 1.x, or Keras with a pre-1.0 Kashgari release.

  2. Install the Kashgari package using the command shown for that selected version range.

  3. For TensorFlow, install the TensorFlow Addons version listed for the applicable TensorFlow release.

  4. Continue with the repository tutorial for the task you intend to build, such as text classification or text labeling.

Expected outcome

A prepared installation path tied to the selected backend, ready to proceed to a task tutorial.

Guide limitations

  • The README specifies Addons pairings for TensorFlow 2.1 through 2.5; it does not give a pairing for every TensorFlow release.
  • This guide covers installation selection and documented next steps, not model training or deployment.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Model inference and serving

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
KashgariTrain a text classifier for labeled text data.
HyperposeCustomize pose-estimation architectures and datasets.Docker, Linux, macOS, Windows
Multi Task NlpDefine multiple tasks and their data and model settings in a YAML file.Linux, macOS, WindowsThe project is an open-source library under the Apache-2.0 license, distributed through its public GitHub repository.
Evidence and transparency

Sources & review

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