Usage

pdist1(P, metric = "euclidean", matrix=False)
pdist2(P, Q, metric = "minkowski", exp = 3)

Arguments:

• two matrices P and Q.

• metric: The distance function to use.

• exp: The exponent of the Minkowski distance.

Installation

The pydist2 library is available on Pypi. Thus, you can install the latest available version using pip:

$pip install pydist2

Supported Python versions

pydist2 has been tested with Python 3.7 and 3.8.

For more information, please checkout the documentation which is available at readthedocs.

This program and the accompanying materials are made available under the terms of the MIT License.

Progress & Features

• [X] Commit the first code’s version.

• [X] Support the following list of distances.

• [X] Display the distance in a matrix form(a combination for each pair of points):

  >>> X = np.array([[100, 100],[0, 100],[100, 0], [500, 400], [300, 600]])
  >>> pdist1(X,matrix=True) # by default, metric = 'euclidean'
  array([[100.    , 100.    , 100.    ,   0.    , 100.    ],
         [100.    , 100.    , 100.    , 100.    ,   0.    ],
         [500.    , 100.    , 100.    , 500.    , 400.    ],
         [538.5165, 100.    , 100.    , 300.    , 600.    ],
         [141.4214,   0.    , 100.    , 100.    ,   0.    ],
         [583.0952,   0.    , 100.    , 500.    , 400.    ],
         [583.0952,   0.    , 100.    , 300.    , 600.    ],
         [565.6854, 100.    ,   0.    , 500.    , 400.    ],
         [632.4555, 100.    ,   0.    , 300.    , 600.    ],
         [282.8427, 500.    , 400.    , 300.    , 600.    ]])

where the first column represents the distance between each pair of observations. for instance, the euclidean distance between (100. , 100.) and ( 0. , 100.) is 100.

• [X] Support numpy arrays of the same size only.

Todo list

• [ ] Re-validate the correctness of the distances equations.

• [ ] Performance tests & vectorization.

• [ ] Adding new distances.

• [ ] Adding a squared form of the distance.

• [ ] Support tuples and list.

• [ ] Write more test cases.

• [ ] Handling Exceptions.

• [ ] Restructure the docs.

0.0.1 (2021-03-04)

• First release on PyPI.

0.0.5 (2021-03-21)

• Adding sqeuclidean metric.