# -*- coding: utf-8 -*-
"""
Created on Fri Nov 10 11:18:31 2017

@author: Shiro
"""
from sklearn.decomposition import PCA
import math
import numpy as np
import cv2
def load_train_from_file(file):
        f = open(file, 'r')
        data = [line.split() for line in f]
        size = cv2.imread(data[0][0], 0).shape
        width = size[1]
        height = size[0]
        Y = np.zeros((len(data), ), dtype=int)
        X = []
        print('Shape Y : ', Y.shape)
        for i, ldata in enumerate(data):
            Y[i] = int(data[i][1])
            X.append(cv2.imread(data[i][0], 0).reshape(-1))
#            cv2.imshow('image', X[i][:].reshape(self.height, self.width))
#            cv2.waitKey(0)
#            cv2.destroyAllWindows()

        
        return np.asarray(X), Y, height, width
    
def load_data_from_file(file):
        f = open(file, 'r')
        data = [line.split() for line in f]
        Y = np.zeros((len(data), ), dtype=int)
        X = []
        print('Shape Y test : ', Y.shape)
        for i, ldata in enumerate(data):
            Y[i] = int(data[i][1])
            X.append(cv2.imread(data[i][0], 0).reshape(-1))
      
        return np.asarray(X), Y
 
def get_model(X_train):
    # Compute a PCA 
    n_components = 9
    
    pca = PCA(n_components=n_components, whiten=True).fit(X_train)
 
    # apply PCA transformation
    X_train_pca = pca.transform(X_train)
    print('X train shape', X_train_pca.shape)
    return pca, X_train_pca

def predict(X_train_pca, y, X_test_pca):

    distances = []
    # Calculate euclidian distance from test image to each of the known images and save distances
    for i, test_pca in enumerate(X_train_pca):
            #print(X_test_pca.shape)
            #print(test_pca.shape)
            dist = math.sqrt(sum([diff**2 for diff in (X_test_pca - test_pca)]))
            distances.append(dist)
 
    found_ID = y[np.argmin(np.asarray(distances))]#min(distances)[1]
    return found_ID

def evaluate(pca, X_train_pca, y, X_test, y_test):
    X_test_pca = pca.transform(X_test)
    tot = len(y_test)
    res = 0.0
    for i, img_test in enumerate(X_test_pca):
    
        pred = int(predict(X_train_pca, y, img_test))
        if pred == y_test[i]:
            res = res + 1
    return res/tot
def __main__():
    X_train, Y_train, height, width = load_train_from_file('train10.txt')
    x_test, y_test = load_data_from_file('test10.txt')
    pca, x_train_pca = get_model(X_train)
    acc = evaluate(pca, x_train_pca, Y_train, x_test, y_test)
    print('accuracy : ', acc)
    
__main__()