py4pnn 0.0.3

Pseudo Nearest Neighbors Python Library

py4pnn

Pseudo Nearest Neighbors Python Library

Getting Started

This project is simply implementation of Pseudo Nearest Neighbors algorithm in python programming language.

Prerequisites

This Project Has No Prerequisites

Installing

The easiest way to install py4pnn is using pip

pip install py4pnn

Usage

There is only 1 public method of pnn class. It is predict method, it takes 5 argument namely x_train, t_train, x_test, k, and distance calculation method. We provide 6 distance namely euclidean, squared_euclidean, manhattan, canberra, chebyshev, and bray_curtis.

from py4pnn.p_nearest_neighbors import pnn
classifier = pnn()
x_train = [[0.23,0.54],[0.45,1.23],[1.54,0.23],[0.93,0.535]]
t_train = [0,1,0,1]
x_test = [[0.34,0.65],[0.90,0.50]]
y = classifier.predict(x_train,t_train,x_test,1,'euclidean')