Python module for Factorial Analysis
Project description
fanalysis
fanalysis is a Python module for Factorial Analysis distributed under the 3-Clause BSD license.
With this fanalysis package, you can perform:
- Simple Correspondence Analysis
- Multiple Correspondence Analysis
- Principal Components Analysis
Those statistical methods can be used in two ways:
- as descriptive methods ("datamining approach")
- as reduction methods in scikit-learn pipelines ("machine learning approach")
Installation
Dependencies
fanalysis requires:
Python 3
NumPy >= 1.11.0
Matplotlib >= 2.0.0
Scikit-learn >= 0.18.0
Pandas >= 0.19.0
User installation
You can install fanalysis using pip:
pip install fanalysis
Running the tests
After installation, you can launch the test suite from outside the source directory:
python -m unittest
The philosophy of the unit tests consists in comparing the outputs of fanalysis (with various combinations of parameters) with the outputs of the R FactoMineR package.
Documentation
The docstring is written in english.
Tutorials are available in french:
https://github.com/OlivierGarciaDev/fanalysis/blob/master/doc/ca_tutorial.ipynb
https://github.com/OlivierGarciaDev/fanalysis/blob/master/doc/mca_tutorial.ipynb
https://github.com/OlivierGarciaDev/fanalysis/blob/master/doc/pca_tutorial.ipynb
Author
Olivier Garcia (o.garcia.dev@gmail.com)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for fanalysis-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 8d43ecf5b6b85869f9ffc9de9c79d7b4d64f87a3204c17687717e7c9863fef49 |
|
MD5 | cba698e3c9d2a0d66b5d9c8639675724 |
|
BLAKE2b-256 | 97b90abf11b0e44fdb450c745e4c9a1e4ab764d5ad8882011161c3ddf290a789 |