Metadata-Version: 2.1
Name: statLM
Version: 0.0.2
Summary: Language models to predict words
Home-page: https://github.com/RaRedmer/statLM
Author: Raphael Redmer
Author-email: ra.redmer@outlook.com
License: MIT
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.7
Description-Content-Type: text/x-rst
Requires-Dist: nose
Requires-Dist: sphinx
Requires-Dist: sklearn

******
statLM
******

statLM (Statistical Language Models) is a library for classical as well as modern language models.

Example Usage
#############

Train a language model and make predictions based on queries i.e. test data.

.. code-block:: python

    corpus = ["let us see were this project leads us",
                "we are having great fun so far",
                "we are actively developing",
                "it is getting tougher but it is still fun",
                "this project teaches us how to construct test cases"] 

    sb = StupidBackoff(n_max=3, alpha=0.4)
    # fit model on corpus
    sb.fit( corpus )
    # make predictions
    queries = ["let us see were that project", "how many options"]
    sb.predict(queries)


In Progress
###########

* more language models
* improve efficiency of ngram comparisons
* construct CD/CI tests via github action
* add type checking

Copyright
#########

Copyright (C) 2020 statLM Raphael Redmer

For license information, see LICENSE.txt.


