https://github.com/thread/wide-product

Fast column-wise partial Khatri-Rao product
https://github.com/thread/wide-product

Last synced: about 2 months ago
JSON representation

Fast column-wise partial Khatri-Rao product

• Host: GitHub
• URL: https://github.com/thread/wide-product
• Owner: thread
• License: mit
• Created: 2017-04-05T12:52:53.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2022-11-18T11:04:09.000Z (about 2 years ago)
• Last Synced: 2024-10-01T14:58:05.640Z (4 months ago)
• Language: Python
• Homepage:
• Size: 36.1 KB
• Stars: 1
• Watchers: 21
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.rst
  • License: COPYING

Awesome Lists containing this project

README

        
wide-product

============

`wide-product` implements a partial, column-wise Khatri-Rao product. It is fast,

and works on sparse matrices.

It can be useful for engineering of cross-features for machine learning.

Definition

----------

For a pair of scalars (~ one by one matrices), the wide product is

multiplication:

.. code:: python

  wide_product ( [[a]], [[b]] ) == [[a * b]]

Where matrices are constructed by *vertical stacking*, the product is row-wise:

.. code:: python

    wide_product ( vstack((A, B)), vstack((C, D)) ) ==

        vstack((wide_product(A, C),

                wide_product(B, D)))

Where matrices are constructed by *horizontal stacking*, the product contains

all the products of the subcomponents up to permutation of columns:

.. code:: python

    wide_product ( hstack((A, B)), hstack((C, D)) ) ==

        hstack((wide_product(A, C),

                wide_product(A, D),

                wide_product(B, C),

                wide_product(B, D)))

Installation

------------

.. code:: bash

  pip install wide-product

Development

-----------

To build the module:

.. code:: bash

  python setup.py build

To test:

.. code:: bash

  PYTHONPATH=$(echo build/lib*):. py.test

To install:

.. code:: bash

  pip install .