# asyml/texar

Toolkit for Machine Learning, Natural Language Processing, and Text Generation, in TensorFlow.  This is part of the CASL project: http://casl-project.ai/

Repository: https://github.com/asyml/texar
Canonical: https://ross.abutalabs.com/products/texar
Homepage: https://asyml.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: machine-learning, natural-language-processing, tensorflow, deep-learning, text-generation, python, machine-translation, dialog-systems, texar, bert, gpt-2, xlnet, text-data, data-processing, casl-project
Last push: 2026-07-21T02:23:58+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 93, release rhythm 8, longevity 100
- inputs: {"age_days": 3329, "days_push": 44, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2389, forks 367 (observed 2026-08-28T04:06:43.247217+00:00)

## What it is
Texar is a modularized Python toolkit for machine learning, especially natural language processing and text generation, built on TensorFlow. It provides reusable modules for data processing, model architectures, training algorithms, and pretrained models like BERT, GPT-2, and XLNet.

## Use cases
- build text generation models in tensorflow
- fine-tune bert for classification
- train sequence-to-sequence machine translation models
- prototype nlp models quickly
- build dialog systems
- run distributed multi-gpu model training
- use pretrained gpt-2 for text generation

## When to choose
- you work in TensorFlow and want modular NLP/text-generation components
- you need uniform interfaces to multiple pretrained models like BERT, GPT-2, XLNet
- you want fast prototyping of encoder-decoder and generative models
- you need both research flexibility and high-level APIs

## When to avoid
- you prefer PyTorch - use Texar-PyTorch instead
- you need actively maintained libraries with recent framework support
- your project targets TensorFlow 2.x or newer ecosystems
- you only need simple off-the-shelf inference without customization

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, nlp, llm-training, data-science
- domain: machine-learning, deep-learning
- platform: python, cross-platform
- tags: tensorflow, text-generation, pretrained-models, bert, gpt-2, xlnet, machine-translation, dialog-systems, natural-language-processing

## Member repositories
- asyml/texar (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:43.247217+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:34:31.447283+00:00, confidence not recorded.
  - readme: https://github.com/asyml/texar (fetched 2026-08-28T04:06:43.247217+00:00, sha e9d1f18da8f6)
  - homepage: https://asyml.io (fetched 2026-08-29T10:15:12.112420+00:00, sha 592fc2255e80)
  - registry_pypi: https://pypi.org/pypi/texar/json (fetched 2026-08-29T10:15:12.121506+00:00, sha fa04669ea05e)
- Data as of 2026-08-30T08:39:29.467469+00:00.