Metadata-Version: 2.1
Name: counterix
Version: 1.0.1
Summary: Generating count-based Distributional Semantic Models
Home-page: https://github.com/akb89/counterix
Author:  Alexandre Kabbach
Author-email: akb@3azouz.net
License: MIT
Download-URL: https://github.com/akb89/counterix
Description: # counterix
        [![GitHub release][release-image]][release-url]
        [![PyPI release][pypi-image]][pypi-url]
        [![Build][build-image]][build-url]
        [![MIT License][license-image]][license-url]
        
        
        [release-image]:https://img.shields.io/github/release/akb89/counterix.svg?style=flat-square
        [release-url]:https://github.com/akb89/counterix/releases/latest
        [pypi-image]:https://img.shields.io/pypi/v/counterix.svg?style=flat-square
        [pypi-url]:https://pypi.org/project/counterix/
        [build-image]:https://img.shields.io/github/workflow/status/akb89/counterix/CI?style=flat-square
        [build-url]:https://github.com/akb89/counterix/actions?query=workflow%3ACI
        [license-image]:http://img.shields.io/badge/license-MIT-000000.svg?style=flat-square
        [license-url]:LICENSE.txt
        
        A small toolkit to generate count-based PPMI-weighed SVD Distributional Semantic Models.
        
        ## Install
        ```shell
        pip install counterix
        ```
        
        or, after a git clone:
        ```shell
        python3 setup.py install
        ```
        
        ## Use
        
        ### Generate
        To generate a raw count matrix from a tokenized corpus, run:
        ```shell
        counterix generate \
          --corpus /abs/path/to/corpus/txt/file \
          --min-count frequency_threshold \
          --win-size window_size
        ```
        
        If the `--output` parameter is not set, the output files will be saved to the corpus directory.
        
        ### Weigh
        To weigh a raw count model with PPMI, run:
        ```
        counterix weigh --model /abs/path/to/raw/count/npz/model
        ```
        
        ### SVD
        To apply SVD on a PPMI-weighed model, with k=10000, run:
        ```
        counterix svd \
          --model /abs/path/to/ppmi/npz/model \
          --dim singular_vectors_final_dim
        ```
        
        To control the number of threads used during SVD, run counterix with `env OMP_NUM_THREADS=1`
        
Keywords: count-based,distributional semantics
Platform: any
Classifier: Development Status :: 5 - Production/Stable
Classifier: Environment :: Web Environment
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Education
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Natural Language :: English
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3.5
Classifier: Programming Language :: Python :: 3.6
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Description-Content-Type: text/markdown
