sklearn-Perceptron.py

'''
Created on Oct 16, 2016

@author: Cheng-lin Li
'''
from __future__ import print_function
import numpy as np
import math
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from sklearn.linear_model import Perceptron

def getInputData(filename):
    _data = np.genfromtxt(filename, delimiter=',')
    _X = _data[:, :3]
    _Y = _data[:, 3]
    return _X, _Y

def getInputData1(filename):
    _data = np.genfromtxt(filename, delimiter=',')
    _X = _data[:, :3]
    _Y = _data[:, 4]
    return _X, _Y

def plot(X, Y, W):
    '''
    Plot data into 3D space for showing classification.
    Plot the 3D plane that segments the datas.
    '''
    fig = plt.figure()
    ax = fig.gca(projection='3d')
    ax.set_xlabel('X Label')
    ax.set_ylabel('Y Label')
    ax.set_zlabel('Z Label')

    _data_set = X
    for i in range(len(_data_set)):
        _x = _data_set[i][1]
        _y = _data_set[i][2]
        _z = _data_set[i][3]
        if Y[i] == 1:
            _c_p = ax.scatter(_x, _y, _z, c='c', marker='^') 
        else:
            _r_p = ax.scatter(_x, _y, _z, c='r', marker='o')

    ax.legend ([_c_p, _r_p], ['Label 1 data','Label -1 data'])
    xx, yy = np.meshgrid(np.arange(-0.2, 1.2, 0.02), np.arange(-0.2, 1.2, 0.02))
    zz = -(W[0] + W[1] * xx + W[2] * yy) / W[3]
    ax.plot_surface(xx, yy, zz, color = 'b', alpha = 0.3) # Plot the segmentation plane
    plt.show()

def plot_error(iterationTimes, errorNumber):
    _x = iterationTimes
    _y = errorNumber
    plt.plot(_x, _y)
    plt.xlabel('Iteration Times')
    plt.ylabel('Error number')
    plt.show()
    # Waiting to implement
    pass        

if __name__ == '__main__':
    
    """
    PLA
    """
    X,Y = getInputData("classification.txt")
    X = np.c_[np.ones(len(X)), np.array(X)] # Coordinates vector of points, which is added '1' at first column.
    pla = Perceptron()
    pla.n_iter = 200
    info = pla.get_params()
    print (info)
    pla = pla.fit(X, Y)
    score = pla.score(X, Y)
    W = pla.coef_
    print('score =', score)
    print ('W=', W )
    plot (X, Y, W[0])
    
    """
    pocket PLA
    """
    iterList = []
    numList = []
    best_score = 0
    W = None
    X,Y = getInputData1("classification.txt") #Get column 5 as Y
    X = np.c_[np.ones(len(X)), np.array(X)] # Coordinates vector of points, which is added '1' at first column.
    pla = Perceptron()
    pla.n_iter = 1
    pla.warm_start = True
    info = pla.get_params()
    print (info)  
    for i in range (0, 7000):  
        pla = pla.fit(X, Y)
        score = pla.score(X, Y)
        ErrorNum = (1-score) * 2000
        iterList.append(i)
        numList.append(ErrorNum)
        if (best_score <= score or i == 0):
            best_score = score
            W = pla.coef_ 
    print('score =', best_score)
    print ('W=', W )
    plot_error (iterList, numList)