https://github.com/fmv1992/data_utilities

Data utilities library focused on machine learning and data analysis.
https://github.com/fmv1992/data_utilities

data-science utility-library

Last synced: 20 days ago
JSON representation

Data utilities library focused on machine learning and data analysis.

• Host: GitHub
• URL: https://github.com/fmv1992/data_utilities
• Owner: fmv1992
• License: gpl-3.0
• Archived: true
• Created: 2017-03-18T13:55:58.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2019-10-12T21:19:39.000Z (about 5 years ago)
• Last Synced: 2024-09-21T12:09:32.541Z (about 2 months ago)
• Topics: data-science, utility-library
• Language: Python
• Homepage:
• Size: 229 KB
• Stars: 5
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: readme.md
  • Changelog: changelog.md
  • License: LICENSE

Awesome Lists containing this project

README

        
[![Build Status](https://travis-ci.org/fmv1992/data_utilities.svg?branch=master)](https://travis-ci.org/fmv1992/data_utilities)

# Data Utilities

This module provides some helper functions and conveniences for working with

data analysis in python.

It depends on:

* Numpy

* Scipy

* Pandas

* Matplotlib

* Seaborn

* Scikit-learn

# Organization and files

    .

    ├── data_utilities

    │   ├── __init__.py

    │   ├── matplotlib_utilities.py

    │   ├── pandas_utilities.py

    │   ├── python_utilities.py

    │   ├── sklearn_utilities

    │   │   ├── grid_search.py

    │   │   └── __init__.py

    │   └── tests

    │       ├── __init__.py

    │       ├── test_matplotlib_utilities.py

    │       ├── test_pandas_utilities.py

    │       ├── test_python_utilities.py

    │       ├── test_sklearn_utilities.py

    │       └── test_support.py

    ├── LICENSE

    ├── MANIFEST.in

    ├── readme.md

    └── setup.py

Each of python's significant data modules has its own set of helper functions.

This module does not intend to create its own API or standards. Instead each of

the utilities module should follow the guidelines and APIs provided by the

parent module.

Note: This is a primitive project. Expect backwards incompatible changes as I

figure out the best way to to develop the utilities.

# What's new

* **Added `sklearn_utilities`**.

* Improved tests customization in `du.test`.

* Greatly improved documentation to `matplotlib_utilities`.

* Greatly expanded `pandas_utilities` functions.

* Improved tests as a whole.

# Development guidelines

* Coding style: [PEP 8](https://www.python.org/dev/peps/pep-0008/) compliant.

* Docstrings: [google docstrings](http://sphinxcontrib-napoleon.readthedocs.io/en/latest/example_google.html).

* Before commiting new versions do a test for different versions of python3:

    * python3.4

    * python3.5

    * python3.6

    * (newer versions)

    * Rationale: even though stability is expected between python versions some

      changes occur. See for instance that on commit v1.2.8 (60573d7) there was

      as unexpected import error on python34 but not on python36.

* Support first the test interface of numpy:

        `python3 -c "import data_utilities as du; du.test()"`

  and then the unittest interface:

        `python3 -m unittest discover -vvv data_utilities/tests`