CAMeL-Lab/camel_tools
A suite of Arabic natural language processing tools developed by the CAMeL Lab at New York University Abu Dhabi.
owo·1w ago·June 8, 2026📋 Changes
- Supported Python version are 3.11-3.14.
- Updated dependencies.
- Updated documentation.
- Bug fixes for `BERTUnfactoredDisambiguator`.
- Deprecated `pretrained_cache` in `BERTUnfactoredDisambiguator`.
- Speed improvements for `simple_word_tokenize`.
owo·9mo ago·September 17, 2025Fixed a couple dependency issues.
owo·1y ago·April 15, 2025Fixed errors in `camel_morphology` and `camel_diac`.
owo·1y ago·August 23, 2024Added Python 3.12 support.
owo·1y ago·August 22, 2024Fixed doc generation issues.
owo·1y ago·August 22, 2024Added Python 3.11 support.
owo·3y ago·March 17, 2023Fixes doc build issues.
owo·3y ago·March 17, 2023📋 Changes
- Fixed an issue where pip tries to install camel-kenlm on Windows.
owo·3y ago·March 17, 2023📋 Changes
- Improved BERT disambiguation accuracy.
- Added support for Python 3.10.
owo·4y ago·May 30, 2022📋 Changes
- Fixed an issue importing Unfactored BERT Disambiguator.
owo·4y ago·May 24, 2022📋 Changes
- Added Unfactored BERT disambiguator component.
- Bug fixes
owo·4y ago·March 29, 2022📋 Changes
- Fixed issue with downloading package catalogue on Google Colab.
owo·4y ago·March 29, 2022📋 Changes
- Removed support for Python 3.6 (only 3.7-3.9 are now supported).
- Implemented a new package manager for fine-grained installation of datasets.
- Fixed GPU support for NER and Sentiment Analysis components.
- Added emoji charsets.
- [simple_word_tokenize](https://camel-tools.readthedocs.io/en/latest/api/tokenizers/word.html) now splits emojis correctly and can optionally split digits.
owo·4y ago·August 12, 2021📋 Changes
- Updated documentation and added more examples
- Morphology improvements and bug fixes
owo·5y ago·November 30, 2020This release adds the `camel_data` command line tool for simplifying downloading data sets.
owo·5y ago·November 15, 2020This update fixes installation issues caused by kenlm dependency.
owo·5y ago·November 10, 2020First official release of CAMeL Tools. See [this post](https://camel-lab.github.io/camel_tools_updates/2020/09/08/camel-tools-release-v1.0.0.html) for more information on this release.
owo·6y ago·June 21, 2020This patch fixes errors in the `almor-msa` builtin database. It also provides a new parameter to `camel_tools.tokenizers.morphological.MorphologicalTokenizer` called `diac` that determines wether output tokens are diacritized or not.
owo·6y ago·January 18, 2020📋 Changes
- Fixed issue where diacritics and other marks broke tokenization.
- Fixed escaping of NOAN replacements in backoff analyses.
- Updated import of Mapping abstract base class for Python 3.
owo·6y ago·December 16, 2019📋 Changes
- Fixed handling of sun letters in CALIMA Star.
- Added ANY keyword as values to certain features in CALIMA Star's reinflector.
owo·6y ago·September 12, 2019Fixed a bug in CALIMA Star reinflector that prevented certain POS variants from being generated.
owo·7y ago·May 16, 2019📋 Changes
- Fixed an issue in CalimaStarReinflector that prevented it from generating analyses.
- Fixed missing defines in the almor-msa database.
owo·7y ago·May 9, 2019📋 Changes
- SimpleDisambiguator has been changed to a more fully featured MLEDisambiguator.
- Added word-boundary and morphological tokenizers.
- Added text normalization utilities.
- Analyzer APIs and Disambiguator APIs have been changed to be more general and output more descriptive named tuple objects.
- Almor-msa database now includes extensions.
- CALIMA Star Analyzer now has in-built caching mode.
- CharMapper objects are now callable (no need to use `map_string()` method).
owo·7y ago·February 26, 2019📋 Changes
- Implemented a simple disambiguation function using pos-lex frequencies.
- `camel_calima_star` has a new option to use simple disambiguation in analysis mode.
- `CalimaStarAnalyzer` now has a new method `analyze_words()` to analyze a list of words.
owo·7y ago·December 18, 2018