Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- import pandas as pd
- import cv2
- import numpy as np
- from Lista02 import FuncoesML as fun
- from sklearn.model_selection import KFold
- from sklearn.neighbors import KNeighborsClassifier
- from sklearn.tree import DecisionTreeClassifier
- from sklearn.naive_bayes import GaussianNB
- from sklearn.svm import SVC
- import time
- from sklearn.linear_model import LogisticRegression
- from sklearn.neural_network import MLPClassifier
- from sklearn.ensemble import RandomForestClassifier
- from sklearn.metrics import recall_score
- from sklearn.metrics import precision_score
- from sklearn.metrics import f1_score
- #lendo a imagem
- print('comecou load images')
- squares = []
- squares = fun.loadFiles('C:/Users/Auricelia/Desktop/DataSetsML/New_shapes_dataset/squares/*.jpg', squares)
- circles = []
- circles = fun.loadFiles('C:/Users/Auricelia/Desktop/DataSetsML/New_shapes_dataset/circles/*.jpg',circles)
- triangles = []
- triangles = fun.loadFiles('C:/Users/Auricelia/Desktop/DataSetsML/New_shapes_dataset/triangles/*.jpg', triangles)
- ellipses = []
- ellipses = fun.loadFiles('C:/Users/Auricelia/Desktop/DataSetsML/New_shapes_dataset/ellipses/*.jpg', ellipses)
- trapezia = []
- trapezia = fun.loadFiles('C:/Users/Auricelia/Desktop/DataSetsML/New_shapes_dataset/trapezia/*.jpg', trapezia)
- rectangles = []
- rectangles = fun.loadFiles('C:/Users/Auricelia/Desktop/DataSetsML/New_shapes_dataset/rectangles/*.jpg', rectangles)
- rhombuses = []
- rhombuses = fun.loadFiles('C:/Users/Auricelia/Desktop/DataSetsML/New_shapes_dataset/rhombuses/*.jpg', rhombuses)
- lines = []
- lines = fun.loadFiles('C:/Users/Auricelia/Desktop/DataSetsML/New_shapes_dataset/lines/*.jpg', lines)
- hexagons = []
- hexagons = fun.loadFiles('C:/Users/Auricelia/Desktop/DataSetsML/New_shapes_dataset/hexagons/*.jpg', hexagons)
- print('terminou load images')
- # Selecionando aleatoriamente 72 imagens de cada classe
- squares_selec, squares_naoselec = fun.seleciona_imagens(squares,72)
- circles_selec, circles_naoselec = fun.seleciona_imagens(circles,72)
- triangles_selec, triangles_naoselec = fun.seleciona_imagens(triangles,72)
- ellipses_selec, ellipses_naoselec = fun.seleciona_imagens(ellipses,72)
- trapezia_selec, trapezia_naoselec = fun.seleciona_imagens(trapezia,72)
- rectangles_selec, rectangles_naoselec = fun.seleciona_imagens(rectangles,72)
- rhombuses_selec, rhombuses_naoselec = fun.seleciona_imagens(rhombuses,72)
- lines_selec, lines_naoselec = fun.seleciona_imagens(lines,72)
- hexagons_selec, hexagons_naoselec = fun.seleciona_imagens(hexagons,72)
- #Salvando em pastas diferentes as imagens para seleção de características e as de teste
- fun.save_images(squares_selec,'C:/Users/Auricelia/Desktop/DataSetsML/Random/RandomSquares/')
- fun.save_images(squares_naoselec,'C:/Users/Auricelia/Desktop/DataSetsML/Not_Random/Squares/')
- fun.save_images(circles_selec,'C:/Users/Auricelia/Desktop/DataSetsML/Random/RandomCircles/')
- fun.save_images(circles_naoselec,'C:/Users/Auricelia/Desktop/DataSetsML/Not_Random/Circles/')
- fun.save_images(triangles_selec,'C:/Users/Auricelia/Desktop/DataSetsML/Random/RandomTriangles/')
- fun.save_images(triangles_naoselec,'C:/Users/Auricelia/Desktop/DataSetsML/Not_Random/Triangles/')
- fun.save_images(ellipses_selec, 'C:/Users/Auricelia/Desktop/DataSetsML/Random/RandomEllipses/')
- fun.save_images(ellipses_naoselec,'C:/Users/Auricelia/Desktop/DataSetsML/Not_Random/Ellipses/')
- fun.save_images(trapezia_selec,'C:/Users/Auricelia/Desktop/DataSetsML/Random/RandomTrapezia/')
- fun.save_images(trapezia_naoselec,'C:/Users/Auricelia/Desktop/DataSetsML/Not_Random/Trapezia/')
- fun.save_images(rectangles_selec,'C:/Users/Auricelia/Desktop/DataSetsML/Random/RandomRectangles/')
- fun.save_images(rectangles_naoselec, 'C:/Users/Auricelia/Desktop/DataSetsML/Not_Random/Rectangles/')
- fun.save_images(rhombuses_selec,'C:/Users/Auricelia/Desktop/DataSetsML/Random/RandomRhombuses/')
- fun.save_images(rhombuses_naoselec, 'C:/Users/Auricelia/Desktop/DataSetsML/Not_Random/Rhombuses/')
- fun.save_images(lines_selec, 'C:/Users/Auricelia/Desktop/DataSetsML/Random/RandomLines/')
- fun.save_images(lines_naoselec, 'C:/Users/Auricelia/Desktop/DataSetsML/Not_Random/Lines/')
- fun.save_images(hexagons_selec, 'C:/Users/Auricelia/Desktop/DataSetsML/Random/RandomHexagons/')
- fun.save_images(hexagons_naoselec, 'C:/Users/Auricelia/Desktop/DataSetsML/Not_Random/Hexagons/')
- # PRE PROCESSING
- #criando cópias de cada uma das pastas para redimensionar as imagens
- #quadrados
- squares16_s = squares_selec.copy()
- squares16_n = squares_naoselec.copy()
- squares32_s = squares_selec.copy()
- squares32_n = squares_naoselec.copy()
- squares64_s = squares_selec.copy()
- squares64_n = squares_naoselec.copy()
- squares128_s = squares_selec.copy()
- squares128_n = squares_naoselec.copy()
- squares16_s = fun.resizeImages(squares16_s,16,16)
- squares16_n = fun.resizeImages(squares16_n,16,16)
- squares32_s = fun.resizeImages(squares32_s,32,32)
- squares32_n = fun.resizeImages(squares32_n,32,32)
- squares64_s = fun.resizeImages(squares64_s,64,64)
- squares64_n = fun.resizeImages(squares64_n,64,64)
- squares128_s = fun.resizeImages(squares128_s,128,128)
- squares128_n = fun.resizeImages(squares128_n,128,128)
- #círculos
- circles16_s = circles_selec.copy()
- circles16_n = circles_naoselec.copy()
- circles32_s = circles_selec.copy()
- circles32_n = circles_naoselec.copy()
- circles64_s = circles_selec.copy()
- circles64_n = circles_naoselec.copy()
- circles128_s = circles_selec.copy()
- circles128_n = circles_naoselec.copy()
- circles16_s = fun.resizeImages(circles16_s,16,16)
- circles16_n = fun.resizeImages(circles16_n,16,16)
- circles32_s = fun.resizeImages(circles32_s,32,32)
- circles32_n = fun.resizeImages(circles32_n,32,32)
- circles64_s = fun.resizeImages(circles64_s,64,64)
- circles64_n = fun.resizeImages(circles64_n,64,64)
- circles128_s = fun.resizeImages(circles128_s,128,128)
- circles128_n = fun.resizeImages(circles128_n,128,128)
- #elipses
- ellipsis16_s = ellipses_selec.copy()
- ellipsis16_n = ellipses_naoselec.copy()
- ellipsis32_s = ellipses_selec.copy()
- ellipsis32_n = ellipses_naoselec.copy()
- ellipsis64_s = ellipses_selec.copy()
- ellipsis64_n = ellipses_naoselec.copy()
- ellipsis128_s = ellipses_selec.copy()
- ellipsis128_n = ellipses_naoselec.copy()
- ellipsis16_s = fun.resizeImages(ellipsis16_s,16,16)
- ellipsis16_n = fun.resizeImages(ellipsis16_n,16,16)
- ellipsis32_s = fun.resizeImages(ellipsis32_s,32,32)
- ellipsis32_n = fun.resizeImages(ellipsis32_n,32,32)
- ellipsis64_s = fun.resizeImages(ellipsis64_s,64,64)
- ellipsis64_n = fun.resizeImages(ellipsis64_n,64,64)
- ellipsis128_s = fun.resizeImages(ellipsis128_s,128,128)
- ellipsis128_n = fun.resizeImages(ellipsis128_n,128,128)
- #hexágonos
- hexagons16_s = hexagons_selec.copy()
- hexagons16_n = hexagons_naoselec.copy()
- hexagons32_s = hexagons_selec.copy()
- hexagons32_n = hexagons_naoselec.copy()
- hexagons64_s = hexagons_selec.copy()
- hexagons64_n = hexagons_naoselec.copy()
- hexagons128_s = hexagons_selec.copy()
- hexagons128_n = hexagons_naoselec.copy()
- hexagons16_s = fun.resizeImages(hexagons16_s,16,16)
- hexagons16_n = fun.resizeImages(hexagons16_n,16,16)
- hexagons32_s = fun.resizeImages(hexagons32_s,32,32)
- hexagons32_n = fun.resizeImages(hexagons32_n,32,32)
- hexagons64_s = fun.resizeImages(hexagons64_s,64,64)
- hexagons64_n = fun.resizeImages(hexagons64_n,64,64)
- hexagons128_s = fun.resizeImages(hexagons128_s,128,128)
- hexagons128_n = fun.resizeImages(hexagons128_n,128,128)
- #linhas
- lines16_s = lines_selec.copy()
- lines16_n = lines_naoselec.copy()
- lines32_s = lines_selec.copy()
- lines32_n = lines_naoselec.copy()
- lines64_s = lines_selec.copy()
- lines64_n = lines_naoselec.copy()
- lines128_s = lines_selec.copy()
- lines128_n = lines_naoselec.copy()
- lines16_s = fun.resizeImages(lines16_s,16,16)
- lines16_n = fun.resizeImages(lines16_n,16,16)
- lines32_s = fun.resizeImages(lines32_s,32,32)
- lines32_n = fun.resizeImages(lines32_n,32,32)
- lines64_s = fun.resizeImages(lines64_s,64,64)
- lines64_n = fun.resizeImages(lines64_n,64,64)
- lines128_s = fun.resizeImages(lines128_s,128,128)
- lines128_n = fun.resizeImages(lines128_n,128,128)
- #retângulos
- rectangles16_s = rectangles_selec.copy()
- rectangles16_n = rectangles_naoselec.copy()
- rectangles32_s = rectangles_selec.copy()
- rectangles32_n = rectangles_naoselec.copy()
- rectangles64_s = rectangles_selec.copy()
- rectangles64_n = rectangles_naoselec.copy()
- rectangles128_s = rectangles_selec.copy()
- rectangles128_n = rectangles_naoselec.copy()
- rectangles16_s = fun.resizeImages(rectangles16_s,16,16)
- rectangles16_n = fun.resizeImages(rectangles16_n,16,16)
- rectangles32_s = fun.resizeImages(rectangles32_s,32,32)
- rectangles32_n = fun.resizeImages(rectangles32_n,32,32)
- rectangles64_s = fun.resizeImages(rectangles64_s,64,64)
- rectangles64_n = fun.resizeImages(rectangles64_n,64,64)
- rectangles128_s = fun.resizeImages(rectangles128_s,128,128)
- rectangles128_n = fun.resizeImages(rectangles128_n,128,128)
- #losangos
- rhombuses16_s = rhombuses_selec.copy()
- rhombuses16_n = rhombuses_naoselec.copy()
- rhombuses32_s = rhombuses_selec.copy()
- rhombuses32_n = rhombuses_naoselec.copy()
- rhombuses64_s = rhombuses_selec.copy()
- rhombuses64_n = rhombuses_naoselec.copy()
- rhombuses128_s = rhombuses_selec.copy()
- rhombuses128_n = rhombuses_naoselec.copy()
- rhombuses16_s = fun.resizeImages(rhombuses16_s,16,16)
- rhombuses16_n = fun.resizeImages(rhombuses16_n,16,16)
- rhombuses32_s = fun.resizeImages(rhombuses32_s,32,32)
- rhombuses32_n = fun.resizeImages(rhombuses32_n,32,32)
- rhombuses64_s = fun.resizeImages(rhombuses64_s,64,64)
- rhombuses64_n = fun.resizeImages(rhombuses64_n,64,64)
- rhombuses128_s = fun.resizeImages(rhombuses128_s,128,128)
- rhombuses128_n = fun.resizeImages(rhombuses128_n,128,128)
- #trapézios
- trapezia16_s = trapezia_selec.copy()
- trapezia16_n = trapezia_naoselec.copy()
- trapezia32_s = trapezia_selec.copy()
- trapezia32_n = trapezia_naoselec.copy()
- trapezia64_s = trapezia_selec.copy()
- trapezia64_n = trapezia_naoselec.copy()
- trapezia128_s = trapezia_selec.copy()
- trapezia128_n = trapezia_naoselec.copy()
- trapezia16_s = fun.resizeImages(trapezia16_s,16,16)
- trapezia16_n = fun.resizeImages(trapezia16_n,16,16)
- trapezia32_s = fun.resizeImages(trapezia32_s,32,32)
- trapezia32_n = fun.resizeImages(trapezia32_n,32,32)
- trapezia64_s = fun.resizeImages(trapezia64_s,64,64)
- trapezia64_n = fun.resizeImages(trapezia64_n,64,64)
- trapezia128_s = fun.resizeImages(trapezia128_s,128,128)
- trapezia128_n = fun.resizeImages(trapezia128_n,128,128)
- #triângulos
- triangles16_s = triangles_selec.copy()
- triangles16_n = triangles_naoselec.copy()
- triangles32_s = triangles_selec.copy()
- triangles32_n = triangles_naoselec.copy()
- triangles64_s = triangles_selec.copy()
- triangles64_n = triangles_naoselec.copy()
- triangles128_s = triangles_selec.copy()
- triangles128_n = triangles_naoselec.copy()
- triangles16_s = fun.resizeImages(triangles16_s,16,16)
- triangles16_n = fun.resizeImages(triangles16_n,16,16)
- triangles32_s = fun.resizeImages(triangles32_s,32,32)
- triangles32_n = fun.resizeImages(triangles32_n,32,32)
- triangles64_s = fun.resizeImages(triangles64_s,64,64)
- triangles64_n = fun.resizeImages(triangles64_n,64,64)
- triangles128_s = fun.resizeImages(triangles128_s,128,128)
- triangles128_n = fun.resizeImages(triangles128_n,128,128)
- #convertendo para níveis de cinza
- squares16_s = fun.grayConversion(squares16_s)
- squares16_n = fun.grayConversion(squares16_n)
- squares32_s = fun.grayConversion(squares32_s)
- squares32_n = fun.grayConversion(squares32_n)
- squares64_s = fun.grayConversion(squares64_s)
- squares64_n = fun.grayConversion(squares64_n)
- squares128_s = fun.grayConversion(squares128_s)
- squares128_n = fun.grayConversion(squares128_n)
- circles16_s = fun.grayConversion(circles16_s)
- circles16_n = fun.grayConversion(circles16_n)
- circles32_s = fun.grayConversion(circles32_s)
- circles32_n = fun.grayConversion(circles32_n)
- circles64_s = fun.grayConversion(circles64_s)
- circles64_n = fun.grayConversion(circles64_n)
- circles128_s = fun.grayConversion(circles128_s)
- circles128_n = fun.grayConversion(circles128_n)
- triangles16_s = fun.grayConversion(triangles16_s)
- triangles16_n = fun.grayConversion(triangles16_n)
- triangles32_s = fun.grayConversion(triangles32_s)
- triangles32_n = fun.grayConversion(triangles32_n)
- triangles64_s = fun.grayConversion(triangles64_s)
- triangles64_n = fun.grayConversion(triangles64_n)
- triangles128_s = fun.grayConversion(triangles128_s)
- triangles128_n = fun.grayConversion(triangles128_n)
- trapezia16_s = fun.grayConversion(trapezia16_s)
- trapezia16_n = fun.grayConversion(trapezia16_n)
- trapezia32_s = fun.grayConversion(trapezia32_s)
- trapezia32_n = fun.grayConversion(trapezia32_n)
- trapezia64_s = fun.grayConversion(trapezia64_s)
- trapezia64_n = fun.grayConversion(trapezia64_n)
- trapezia128_s = fun.grayConversion(trapezia128_s)
- trapezia128_n = fun.grayConversion(trapezia128_n)
- rhombuses16_s = fun.grayConversion(rhombuses16_s)
- rhombuses16_n = fun.grayConversion(rhombuses16_n)
- rhombuses32_s = fun.grayConversion(rhombuses32_s)
- rhombuses32_n = fun.grayConversion(rhombuses32_n)
- rhombuses64_s = fun.grayConversion(rhombuses64_s)
- rhombuses64_n = fun.grayConversion(rhombuses64_n)
- rhombuses128_s = fun.grayConversion(rhombuses128_s)
- rhombuses128_n = fun.grayConversion(rhombuses128_n)
- rectangles16_s = fun.grayConversion(rectangles16_s)
- rectangles16_n = fun.grayConversion(rectangles16_n)
- rectangles32_s = fun.grayConversion(rectangles32_s)
- rectangles32_n = fun.grayConversion(rectangles32_n)
- rectangles64_s = fun.grayConversion(rectangles64_s)
- rectangles64_n = fun.grayConversion(rectangles64_n)
- rectangles128_s = fun.grayConversion(rectangles128_s)
- rectangles128_n = fun.grayConversion(rectangles128_n)
- lines16_s = fun.grayConversion(lines16_s)
- lines16_n = fun.grayConversion(lines16_n)
- lines32_s = fun.grayConversion(lines32_s)
- lines32_n = fun.grayConversion(lines32_n)
- lines64_s = fun.grayConversion(lines64_s)
- lines64_n = fun.grayConversion(lines64_n)
- lines128_s = fun.grayConversion(lines128_s)
- lines128_n = fun.grayConversion(lines128_n)
- hexagons16_s = fun.grayConversion(hexagons16_s)
- hexagons16_n = fun.grayConversion(hexagons16_n)
- hexagons32_s = fun.grayConversion(hexagons32_s)
- hexagons32_n = fun.grayConversion(hexagons32_n)
- hexagons64_s = fun.grayConversion(hexagons64_s)
- hexagons64_n = fun.grayConversion(hexagons64_n)
- hexagons128_s = fun.grayConversion(hexagons128_s)
- hexagons128_n = fun.grayConversion(hexagons128_n)
- ellipsis16_s = fun.grayConversion(ellipsis16_s)
- ellipsis16_n = fun.grayConversion(ellipsis16_n)
- ellipsis32_s = fun.grayConversion(ellipsis32_s)
- ellipsis32_n = fun.grayConversion(ellipsis32_n)
- ellipsis64_s = fun.grayConversion(ellipsis64_s)
- ellipsis64_n = fun.grayConversion(ellipsis64_n)
- ellipsis128_s = fun.grayConversion(ellipsis128_s)
- ellipsis128_n = fun.grayConversion(ellipsis128_n)
- #aplicando o filtro gaussiano
- squares16_s = fun.blurConversion(squares16_s,5,0)
- squares16_n = fun.blurConversion(squares16_n,5,0)
- squares32_s = fun.blurConversion(squares32_s,5,0)
- squares32_n = fun.blurConversion(squares32_n,5,0)
- squares64_s = fun.blurConversion(squares64_s,5,0)
- squares64_n = fun.blurConversion(squares64_n,5,0)
- squares128_s = fun.blurConversion(squares128_s,5,0)
- squares128_n = fun.blurConversion(squares128_n,5,0)
- circles16_s = fun.blurConversion(circles16_s, 5, 0)
- circles16_n = fun.blurConversion(circles16_n, 5, 0)
- circles32_s = fun.blurConversion(circles32_s,5 ,0)
- circles32_n = fun.blurConversion(circles32_n,5 ,0)
- circles64_s = fun.blurConversion(circles64_s,5,0)
- circles64_n = fun.blurConversion(circles64_n,5,0)
- circles128_s = fun.blurConversion(circles128_s,5,0)
- circles128_n = fun.blurConversion(circles128_n,5,0)
- triangles16_s = fun.blurConversion(triangles16_s,5,0)
- triangles16_n = fun.blurConversion(triangles16_n,5,0)
- triangles32_s = fun.blurConversion(triangles32_s,5,0)
- triangles32_n = fun.blurConversion(triangles32_n,5,0)
- triangles64_s = fun.blurConversion(triangles64_s,5,0)
- triangles64_n = fun.blurConversion(triangles64_n,5,0)
- triangles128_s = fun.blurConversion(triangles128_s,5,0)
- triangles128_n = fun.blurConversion(triangles128_n,5,0)
- trapezia16_s = fun.blurConversion(trapezia16_s,5,0)
- trapezia16_n = fun.blurConversion(trapezia16_n,5,0)
- trapezia32_s = fun.blurConversion(trapezia32_s,5,0)
- trapezia64_s = fun.blurConversion(trapezia64_s,5,0)
- trapezia64_n = fun.blurConversion(trapezia64_n,5,0)
- trapezia128_s = fun.blurConversion(trapezia128_s,5,0)
- trapezia128_n = fun.blurConversion(trapezia128_n,5,0)
- rhombuses16_s = fun.blurConversion(rhombuses16_s,5,0)
- rhombuses16_n = fun.blurConversion(rhombuses16_n,5,0)
- rhombuses32_s = fun.blurConversion(rhombuses32_s,5,0)
- rhombuses32_n = fun.blurConversion(rhombuses32_n,5,0)
- rhombuses64_s = fun.blurConversion(rhombuses64_s,5,0)
- rhombuses64_n = fun.blurConversion(rhombuses64_n,5,0)
- rhombuses128_s = fun.blurConversion(rhombuses128_s,5,0)
- rhombuses128_n = fun.blurConversion(rhombuses128_n,5,0)
- rectangles16_s = fun.blurConversion(rectangles16_s,5,0)
- rectangles16_n = fun.blurConversion(rectangles16_n,5,0)
- rectangles32_s = fun.blurConversion(rectangles32_s,5,0)
- rectangles32_n = fun.blurConversion(rectangles32_n,5,0)
- rectangles64_s = fun.blurConversion(rectangles64_s,5,0)
- rectangles64_n = fun.blurConversion(rectangles64_n,5,0)
- rectangles128_s = fun.blurConversion(rectangles128_s,5,0)
- rectangles128_n = fun.blurConversion(rectangles128_n,5,0)
- lines16_s = fun.blurConversion(lines16_s,5,0)
- lines16_n = fun.blurConversion(lines16_n,5,0)
- lines32_s = fun.blurConversion(lines32_s,5,0)
- lines32_n = fun.blurConversion(lines32_n,5,0)
- lines64_s = fun.blurConversion(lines64_s,5,0)
- lines64_n = fun.blurConversion(lines64_n,5,0)
- lines128_s = fun.blurConversion(lines128_s,5,0)
- lines128_n = fun.blurConversion(lines128_n,5,0)
- hexagons16_s = fun.blurConversion(hexagons16_s,5,0)
- hexagons16_n = fun.blurConversion(hexagons16_n,5,0)
- hexagons32_s = fun.blurConversion(hexagons32_s,5,0)
- hexagons32_n = fun.blurConversion(hexagons32_n,5,0)
- hexagons64_s = fun.blurConversion(hexagons64_s,5,0)
- hexagons64_n = fun.blurConversion(hexagons64_n,5,0)
- hexagons128_s = fun.blurConversion(hexagons128_s,5,0)
- hexagons128_n = fun.blurConversion(hexagons128_n,5,0)
- ellipsis16_s = fun.blurConversion(ellipsis16_s,5,0)
- ellipsis16_n = fun.blurConversion(ellipsis16_n,5,0)
- ellipsis32_s = fun.blurConversion(ellipsis32_s,5,0)
- ellipsis32_n = fun.blurConversion(ellipsis32_n,5,0)
- ellipsis64_s = fun.blurConversion(ellipsis64_s,5,0)
- ellipsis64_n = fun.blurConversion(ellipsis64_n,5,0)
- ellipsis128_s = fun.blurConversion(ellipsis128_s,5,0)
- ellipsis128_n = fun.blurConversion(ellipsis128_n,5,0)
- #convertendo para binária
- squares16_s = fun.binaryConversion(squares16_s,255,31)
- squares16_n = fun.binaryConversion(squares16_n,255,31)
- squares32_s = fun.binaryConversion(squares32_s,255,31)
- squares32_n = fun.binaryConversion(squares32_n,255,31)
- squares64_s = fun.binaryConversion(squares64_s,255,31)
- squares64_n = fun.binaryConversion(squares64_n,255,31)
- squares128_s = fun.binaryConversion(squares128_s,255,31)
- squares128_n = fun.binaryConversion(squares128_n,255,31)
- circles16_s = fun.binaryConversion(circles16_s, 255, 31)
- circles16_n = fun.binaryConversion(circles16_n, 255, 31)
- circles32_s = fun.binaryConversion(circles32_s,255,31)
- circles32_n = fun.binaryConversion(circles32_n,255,31)
- circles64_s = fun.binaryConversion(circles64_s,255,31)
- circles64_n = fun.binaryConversion(circles64_n,255,31)
- circles128_s = fun.binaryConversion(circles128_s,255,31)
- circles128_n = fun.binaryConversion(circles128_n,255,31)
- triangles16_s = fun.binaryConversion(triangles16_s,255,31)
- triangles16_n = fun.binaryConversion(triangles16_n,255,31)
- triangles32_s = fun.binaryConversion(triangles32_s,255,31)
- triangles32_n = fun.binaryConversion(triangles32_n,255,31)
- triangles64_s = fun.binaryConversion(triangles64_s,255,31)
- triangles64_n = fun.binaryConversion(triangles64_n,255,31)
- triangles128_s = fun.binaryConversion(triangles128_s,255,31)
- triangles128_n = fun.binaryConversion(triangles128_n,255,31)
- trapezia16_s = fun.binaryConversion(trapezia16_s,255,31)
- trapezia16_n = fun.binaryConversion(trapezia16_n,255,31)
- trapezia32_s = fun.binaryConversion(trapezia32_s,255,31)
- trapezia32_n = fun.binaryConversion(trapezia32_n,255,31)
- trapezia64_s = fun.binaryConversion(trapezia64_s,255,31)
- trapezia64_n = fun.binaryConversion(trapezia64_n,255,31)
- trapezia128_s = fun.binaryConversion(trapezia128_s,255,31)
- trapezia128_n = fun.binaryConversion(trapezia128_n,255,31)
- rhombuses16_s = fun.binaryConversion(rhombuses16_s,255,31)
- rhombuses16_n = fun.binaryConversion(rhombuses16_n,255,31)
- rhombuses32_s = fun.binaryConversion(rhombuses32_s,255,31)
- rhombuses32_n = fun.binaryConversion(rhombuses32_n,255,31)
- rhombuses64_s = fun.binaryConversion(rhombuses64_s,255,31)
- rhombuses64_n = fun.binaryConversion(rhombuses64_n,255,31)
- rhombuses128_s = fun.binaryConversion(rhombuses128_s,255,31)
- rhombuses128_n = fun.binaryConversion(rhombuses128_n,255,31)
- rectangles16_s = fun.binaryConversion(rectangles16_s,255,31)
- rectangles16_n = fun.binaryConversion(rectangles16_n,255,31)
- rectangles32_s = fun.binaryConversion(rectangles32_s,255,31)
- rectangles32_n = fun.binaryConversion(rectangles32_n,255,31)
- rectangles64_s = fun.binaryConversion(rectangles64_s,255,31)
- rectangles64_n = fun.binaryConversion(rectangles64_n,255,31)
- rectangles128_s = fun.binaryConversion(rectangles128_s,255,31)
- rectangles128_n = fun.binaryConversion(rectangles128_n,255,31)
- lines16_s = fun.binaryConversion(lines16_s,255,31)
- lines16_n = fun.binaryConversion(lines16_n,255,31)
- lines32_s = fun.binaryConversion(lines32_s,255,31)
- lines32_n = fun.binaryConversion(lines32_n,255,31)
- lines64_s = fun.binaryConversion(lines64_s,255,31)
- lines64_n = fun.binaryConversion(lines64_n,255,31)
- lines128_s = fun.binaryConversion(lines128_s,255,31)
- lines128_n = fun.binaryConversion(lines128_n,255,31)
- hexagons16_s = fun.binaryConversion(hexagons16_s,255,31)
- hexagons16_n = fun.binaryConversion(hexagons16_n,255,31)
- hexagons32_s = fun.binaryConversion(hexagons32_s,255,31)
- hexagons32_n = fun.binaryConversion(hexagons32_n,255,31)
- hexagons64_s = fun.binaryConversion(hexagons64_s,255,31)
- hexagons64_n = fun.binaryConversion(hexagons64_n,255,31)
- hexagons128_s = fun.binaryConversion(hexagons128_s,255,31)
- hexagons128_n = fun.binaryConversion(hexagons128_n,255,31)
- ellipsis16_s = fun.binaryConversion(ellipsis16_s,255,31)
- ellipsis16_n = fun.binaryConversion(ellipsis16_n,255,31)
- ellipsis32_s = fun.binaryConversion(ellipsis32_s,255,31)
- ellipsis32_n = fun.binaryConversion(ellipsis32_n,255,31)
- ellipsis64_s = fun.binaryConversion(ellipsis64_s,255,31)
- ellipsis64_n = fun.binaryConversion(ellipsis64_n,255,31)
- ellipsis128_s = fun.binaryConversion(ellipsis128_s,255,31)
- ellipsis128_n = fun.binaryConversion(ellipsis128_n,255,31)
- #invertendo as cores
- squares16_s = fun.invertConversion(squares16_s)
- squares16_n = fun.invertConversion(squares16_n)
- squares32_s = fun.invertConversion(squares32_s)
- squares32_n = fun.invertConversion(squares32_n)
- squares64_s = fun.invertConversion(squares64_s)
- squares64_n = fun.invertConversion(squares64_n)
- squares128_s = fun.invertConversion(squares128_s)
- squares128_n = fun.invertConversion(squares128_n)
- circles16_s = fun.invertConversion(circles16_s)
- circles16_n = fun.invertConversion(circles16_n)
- circles32_s = fun.invertConversion(circles32_s)
- circles32_n = fun.invertConversion(circles32_n)
- circles64_s = fun.invertConversion(circles64_s)
- circles64_n = fun.invertConversion(circles64_n)
- circles128_s = fun.invertConversion(circles128_s)
- circles128_n = fun.invertConversion(circles128_n)
- triangles16_s = fun.invertConversion(triangles16_s)
- triangles16_n = fun.invertConversion(triangles16_n)
- triangles32_s = fun.invertConversion(triangles32_s)
- triangles32_n = fun.invertConversion(triangles32_n)
- triangles64_s = fun.invertConversion(triangles64_s)
- triangles64_n = fun.invertConversion(triangles64_n)
- triangles128_s = fun.invertConversion(triangles128_s)
- triangles128_n = fun.invertConversion(triangles128_n)
- trapezia16_s = fun.invertConversion(trapezia16_s)
- trapezia16_n = fun.invertConversion(trapezia16_n)
- trapezia32_s = fun.invertConversion(trapezia32_s)
- trapezia32_n = fun.invertConversion(trapezia32_n)
- trapezia64_s = fun.invertConversion(trapezia64_s)
- trapezia64_n = fun.invertConversion(trapezia64_n)
- trapezia128_s = fun.invertConversion(trapezia128_s)
- trapezia128_n = fun.invertConversion(trapezia128_n)
- rhombuses16_s = fun.invertConversion(rhombuses16_s)
- rhombuses16_n = fun.invertConversion(rhombuses16_n)
- rhombuses32_s = fun.invertConversion(rhombuses32_s)
- rhombuses32_n = fun.invertConversion(rhombuses32_n)
- rhombuses64_s = fun.invertConversion(rhombuses64_s)
- rhombuses64_n = fun.invertConversion(rhombuses64_n)
- rhombuses128_s = fun.invertConversion(rhombuses128_s)
- rhombuses128_n = fun.invertConversion(rhombuses128_n)
- rectangles16_s = fun.invertConversion(rectangles16_s)
- rectangles16_n = fun.invertConversion(rectangles16_n)
- rectangles32_s = fun.invertConversion(rectangles32_s)
- rectangles32_n = fun.invertConversion(rectangles32_n)
- rectangles64_s = fun.invertConversion(rectangles64_s)
- rectangles64_n = fun.invertConversion(rectangles64_n)
- rectangles128_s = fun.invertConversion(rectangles128_s)
- rectangles128_n = fun.invertConversion(rectangles128_n)
- lines16_s = fun.invertConversion(lines16_s)
- lines16_n = fun.invertConversion(lines16_n)
- lines32_s = fun.invertConversion(lines32_s)
- lines32_n = fun.invertConversion(lines32_n)
- lines64_s = fun.invertConversion(lines64_s)
- lines64_n = fun.invertConversion(lines64_n)
- lines128_s = fun.invertConversion(lines128_s)
- lines128_n = fun.invertConversion(lines128_n)
- hexagons16_s = fun.invertConversion(hexagons16_s)
- hexagons16_n = fun.invertConversion(hexagons16_n)
- hexagons32_s = fun.invertConversion(hexagons32_s)
- hexagons32_n = fun.invertConversion(hexagons32_n)
- hexagons64_s = fun.invertConversion(hexagons64_s)
- hexagons64_n = fun.invertConversion(hexagons64_n)
- hexagons128_s = fun.invertConversion(hexagons128_s)
- hexagons128_n = fun.invertConversion(hexagons128_n)
- ellipsis16_s = fun.invertConversion(ellipsis16_s)
- ellipsis16_n = fun.invertConversion(ellipsis16_n)
- ellipsis32_s = fun.invertConversion(ellipsis32_s)
- ellipsis32_n = fun.invertConversion(ellipsis32_n)
- ellipsis64_s = fun.invertConversion(ellipsis64_s)
- ellipsis64_n = fun.invertConversion(ellipsis64_n)
- ellipsis128_s = fun.invertConversion(ellipsis128_s)
- ellipsis128_n = fun.invertConversion(ellipsis128_n)
- print('terminou pre processing')
- # extraindo caracteristicas das imagens
- squares128_vector_s = fun.extratorCaracteristicas(squares128_s)
- squares128_vector_n = fun.extratorCaracteristicas(squares128_n)
- circles128_vector_s = fun.extratorCaracteristicas(circles128_s)
- circles128_vector_n = fun.extratorCaracteristicas(circles128_n)
- triangles128_vector_s = fun.extratorCaracteristicas(triangles128_s)
- triangles128_vector_n = fun.extratorCaracteristicas(triangles128_n)
- trapezia128_vector_s = fun.extratorCaracteristicas(trapezia128_s)
- trapezia128_vector_n = fun.extratorCaracteristicas(trapezia128_n)
- rhombuses128_vector_s = fun.extratorCaracteristicas(rhombuses128_s)
- rhombuses128_vector_n = fun.extratorCaracteristicas(rhombuses128_n)
- rectangles128_vector_s = fun.extratorCaracteristicas(rectangles128_s)
- rectangles128_vector_n = fun.extratorCaracteristicas(rectangles128_n)
- lines128_vector_s = fun.extratorCaracteristicas(lines128_s)
- lines128_vector_n = fun.extratorCaracteristicas(lines128_n)
- hexagons128_vector_s = fun.extratorCaracteristicas(hexagons128_s)
- hexagons128_vector_n = fun.extratorCaracteristicas(hexagons128_n)
- ellipsis128_vector_s = fun.extratorCaracteristicas(ellipsis128_s)
- ellipsis128_vector_n = fun.extratorCaracteristicas(ellipsis128_n)
- squares64_vector_s = fun.extratorCaracteristicas(squares64_s)
- squares64_vector_n = fun.extratorCaracteristicas(squares64_n)
- circles64_vector_s = fun.extratorCaracteristicas(circles64_s)
- circles64_vector_n = fun.extratorCaracteristicas(circles64_n)
- triangles64_vector_s = fun.extratorCaracteristicas(triangles64_s)
- triangles64_vector_n = fun.extratorCaracteristicas(triangles64_n)
- trapezia64_vector_s = fun.extratorCaracteristicas(trapezia64_s)
- trapezia64_vector_n = fun.extratorCaracteristicas(trapezia64_n)
- rhombuses64_vector_s = fun.extratorCaracteristicas(rhombuses64_s)
- rhombuses64_vector_n = fun.extratorCaracteristicas(rhombuses64_n)
- rectangles64_vector_s = fun.extratorCaracteristicas(rectangles64_s)
- rectangles64_vector_n = fun.extratorCaracteristicas(rectangles64_n)
- lines64_vector_s = fun.extratorCaracteristicas(lines64_s)
- lines64_vector_n = fun.extratorCaracteristicas(lines64_n)
- hexagons64_vector_s = fun.extratorCaracteristicas(hexagons64_s)
- hexagons64_vector_n = fun.extratorCaracteristicas(hexagons64_n)
- ellipsis64_vector_s = fun.extratorCaracteristicas(ellipsis64_s)
- ellipsis64_vector_n = fun.extratorCaracteristicas(ellipsis64_n)
- squares32_vector_s = fun.extratorCaracteristicas(squares32_s)
- squares32_vector_n = fun.extratorCaracteristicas(squares32_n)
- circles32_vector_s = fun.extratorCaracteristicas(circles32_s)
- circles32_vector_n = fun.extratorCaracteristicas(circles32_n)
- triangles32_vector_s = fun.extratorCaracteristicas(triangles32_s)
- triangles32_vector_n = fun.extratorCaracteristicas(triangles32_n)
- trapezia32_vector_s = fun.extratorCaracteristicas(trapezia32_s)
- trapezia32_vector_n = fun.extratorCaracteristicas(trapezia32_n)
- rhombuses32_vector_s = fun.extratorCaracteristicas(rhombuses32_s)
- rhombuses32_vector_n = fun.extratorCaracteristicas(rhombuses32_n)
- rectangles32_vector_s = fun.extratorCaracteristicas(rectangles32_s)
- rectangles32_vector_n = fun.extratorCaracteristicas(rectangles32_n)
- lines32_vector_s = fun.extratorCaracteristicas(lines32_s)
- lines32_vector_n = fun.extratorCaracteristicas(lines32_n)
- hexagons32_vector_s = fun.extratorCaracteristicas(hexagons32_s)
- hexagons32_vector_n = fun.extratorCaracteristicas(hexagons32_n)
- ellipsis32_vector_s = fun.extratorCaracteristicas(ellipsis32_s)
- ellipsis32_vector_n = fun.extratorCaracteristicas(ellipsis32_n)
- squares16_vector_s = fun.extratorCaracteristicas(squares16_s)
- squares16_vector_n = fun.extratorCaracteristicas(squares16_n)
- circles16_vector_s = fun.extratorCaracteristicas(circles16_s)
- circles16_vector_n = fun.extratorCaracteristicas(circles16_n)
- triangles16_vector_s = fun.extratorCaracteristicas(triangles16_s)
- triangles16_vector_n = fun.extratorCaracteristicas(triangles16_n)
- trapezia16_vector_s = fun.extratorCaracteristicas(trapezia16_s)
- trapezia16_vector_n = fun.extratorCaracteristicas(trapezia16_n)
- rhombuses16_vector_s = fun.extratorCaracteristicas(rhombuses16_s)
- rhombuses16_vector_n = fun.extratorCaracteristicas(rhombuses16_n)
- rectangles16_vector_s = fun.extratorCaracteristicas(rectangles16_s)
- rectangles16_vector_n = fun.extratorCaracteristicas(rectangles16_n)
- lines16_vector_s = fun.extratorCaracteristicas(lines16_s)
- lines16_vector_n = fun.extratorCaracteristicas(lines16_n)
- hexagons16_vector_s = fun.extratorCaracteristicas(hexagons16_s)
- hexagons16_vector_n = fun.extratorCaracteristicas(hexagons16_n)
- ellipsis16_vector_s = fun.extratorCaracteristicas(ellipsis16_s)
- ellipsis16_vector_n = fun.extratorCaracteristicas(ellipsis16_n)
- print('terminou extracao carac')
- # transformando os vetores em dataframes
- squares128_vector_s = pd.DataFrame(squares128_vector_s)
- squares128_vector_n = pd.DataFrame(squares128_vector_n)
- circles128_vector_s = pd.DataFrame(circles128_vector_s)
- circles128_vector_n = pd.DataFrame(circles128_vector_n)
- triangles128_vector_s = pd.DataFrame(triangles128_vector_s)
- triangles128_vector_n = pd.DataFrame(triangles128_vector_n)
- trapezia128_vector_s = pd.DataFrame(trapezia128_vector_s)
- trapezia128_vector_n = pd.DataFrame(trapezia128_vector_n)
- rhombuses128_vector_s = pd.DataFrame(rhombuses128_vector_s)
- rhombuses128_vector_n = pd.DataFrame(rhombuses128_vector_n)
- rectangles128_vector_s = pd.DataFrame(rectangles128_vector_s)
- rectangles128_vector_n = pd.DataFrame(rectangles128_vector_n)
- lines128_vector_s = pd.DataFrame(lines128_vector_s)
- lines128_vector_n = pd.DataFrame(lines128_vector_n)
- hexagons128_vector_s = pd.DataFrame(hexagons128_vector_s)
- hexagons128_vector_n = pd.DataFrame(hexagons128_vector_n)
- ellipsis128_vector_s = pd.DataFrame(ellipsis128_vector_s)
- ellipsis128_vector_n = pd.DataFrame(ellipsis128_vector_n)
- squares32_vector_s = pd.DataFrame(squares32_vector_s)
- squares32_vector_n = pd.DataFrame(squares32_vector_n)
- circles32_vector_s = pd.DataFrame(circles32_vector_s)
- circles32_vector_n = pd.DataFrame(circles32_vector_n)
- triangles32_vector_s = pd.DataFrame(triangles32_vector_s)
- triangles32_vector_n = pd.DataFrame(triangles32_vector_n)
- trapezia32_vector_s = pd.DataFrame(trapezia32_vector_s)
- trapezia32_vector_n = pd.DataFrame(trapezia32_vector_n)
- rhombuses32_vector_s = pd.DataFrame(rhombuses32_vector_s)
- rhombuses32_vector_n = pd.DataFrame(rhombuses32_vector_n)
- rectangles32_vector_s = pd.DataFrame(rectangles32_vector_s)
- rectangles32_vector_n = pd.DataFrame(rectangles32_vector_n)
- hexagons32_vector_s = pd.DataFrame(hexagons32_vector_s)
- hexagons32_vector_n = pd.DataFrame(hexagons32_vector_n)
- ellipsis32_vector_s = pd.DataFrame(ellipsis32_vector_s)
- ellipsis32_vector_n = pd.DataFrame(ellipsis32_vector_n)
- lines32_vector_s = pd.DataFrame(lines32_vector_s)
- lines32_vector_n = pd.DataFrame(lines32_vector_n)
- squares64_vector_s = pd.DataFrame(squares64_vector_s)
- squares64_vector_n = pd.DataFrame(squares64_vector_n)
- circles64_vector_s = pd.DataFrame(circles64_vector_s)
- circles64_vector_n = pd.DataFrame(circles64_vector_n)
- triangles64_vector_s = pd.DataFrame(triangles64_vector_s)
- triangles64_vector_n = pd.DataFrame(triangles64_vector_n)
- trapezia64_vector_s = pd.DataFrame(trapezia64_vector_s)
- trapezia64_vector_n = pd.DataFrame(trapezia64_vector_n)
- rhombuses64_vector_s = pd.DataFrame(rhombuses64_vector_s)
- rhombuses64_vector_n = pd.DataFrame(rhombuses64_vector_n)
- rectangles64_vector_s = pd.DataFrame(rectangles64_vector_s)
- rectangles64_vector_n = pd.DataFrame(rectangles64_vector_n)
- lines64_vector_s = pd.DataFrame(lines64_vector_s)
- lines64_vector_n = pd.DataFrame(lines64_vector_n)
- hexagons64_vector_s = pd.DataFrame(hexagons64_vector_s)
- hexagons64_vector_n = pd.DataFrame(hexagons64_vector_n)
- ellipsis64_vector_s = pd.DataFrame(ellipsis64_vector_s)
- ellipsis64_vector_n = pd.DataFrame(ellipsis64_vector_n)
- circles16_vector_s = pd.DataFrame(circles16_vector_s)
- circles16_vector_n = pd.DataFrame(circles16_vector_n)
- squares16_vector_s = pd.DataFrame(squares16_vector_s)
- squares16_vector_n = pd.DataFrame(squares16_vector_n)
- triangles16_vector_s = pd.DataFrame(triangles16_vector_s)
- triangles16_vector_n = pd.DataFrame(triangles16_vector_n)
- trapezia16_vector_s = pd.DataFrame(trapezia16_vector_s)
- trapezia16_vector_n = pd.DataFrame(trapezia16_vector_n)
- rhombuses16_vector_s = pd.DataFrame(rhombuses16_vector_s)
- rhombuses16_vector_n = pd.DataFrame(rhombuses16_vector_n)
- rectangles16_vector_s = pd.DataFrame(rectangles16_vector_s)
- rectangles16_vector_n = pd.DataFrame(rectangles16_vector_n)
- lines16_vector_s = pd.DataFrame(lines16_vector_s)
- lines16_vector_n = pd.DataFrame(lines16_vector_n)
- hexagons16_vector_s = pd.DataFrame(hexagons16_vector_s)
- hexagons16_vector_n = pd.DataFrame(hexagons16_vector_n)
- ellipsis16_vector_s = pd.DataFrame(ellipsis16_vector_s)
- ellipsis16_vector_n = pd.DataFrame(ellipsis16_vector_n)
- print('terminou transformar em dataframe')
- #incluindo a classe nos dataframes
- squares128_vector_s['Classe'] = 'square'
- squares128_vector_n['Classe'] = 'square'
- circles128_vector_s['Classe'] = 'circle'
- circles128_vector_n['Classe'] = 'circle'
- triangles128_vector_s['Classe'] = 'triangle'
- triangles128_vector_n['Classe'] = 'triangle'
- trapezia128_vector_s['Classe'] = 'trapezia'
- trapezia128_vector_n['Classe'] = 'trapezia'
- rhombuses128_vector_s['Classe'] = 'rhombuse'
- rhombuses128_vector_n['Classe'] = 'rhombuse'
- rectangles128_vector_s['Classe'] = 'rectangle'
- rectangles128_vector_n['Classe'] = 'rectangle'
- lines128_vector_s['Classe'] = 'line'
- lines128_vector_n['Classe'] = 'line'
- hexagons128_vector_s['Classe'] = 'hexagon'
- hexagons128_vector_n['Classe'] = 'hexagon'
- ellipsis128_vector_s['Classe'] = 'ellipse'
- ellipsis128_vector_n['Classe'] = 'ellipse'
- squares32_vector_s['Classe'] = 'square'
- squares32_vector_n['Classe'] = 'square'
- circles32_vector_s['Classe'] = 'circle'
- circles32_vector_n['Classe'] = 'circle'
- triangles32_vector_s['Classe'] = 'triangle'
- triangles32_vector_n['Classe'] = 'triangle'
- trapezia32_vector_s['Classe'] = 'trapezia'
- trapezia32_vector_n['Classe'] = 'trapezia'
- rhombuses32_vector_s['Classe'] = 'rhombuse'
- rhombuses32_vector_n['Classe'] = 'rhombuse'
- rectangles32_vector_s['Classe'] = 'rectangle'
- rectangles32_vector_n['Classe'] = 'rectangle'
- lines32_vector_s['Classe'] = 'line'
- lines32_vector_n['Classe'] = 'line'
- hexagons32_vector_s['Classe'] = 'hexagon'
- hexagons32_vector_n['Classe'] = 'hexagon'
- ellipsis32_vector_s['Classe'] = 'ellipse'
- ellipsis32_vector_n['Classe'] = 'ellipse'
- squares64_vector_s['Classe'] = 'square'
- squares64_vector_n['Classe'] = 'square'
- circles64_vector_s['Classe'] = 'circle'
- circles64_vector_n['Classe'] = 'circle'
- triangles64_vector_s['Classe'] = 'triangle'
- triangles64_vector_n['Classe'] = 'triangle'
- trapezia64_vector_s['Classe'] = 'trapezia'
- trapezia64_vector_n['Classe'] = 'trapezia'
- rhombuses64_vector_s['Classe'] = 'rhombuse'
- rhombuses64_vector_n['Classe'] = 'rhombuse'
- rectangles64_vector_s['Classe'] = 'rectangle'
- rectangles64_vector_n['Classe'] = 'rectangle'
- lines64_vector_s['Classe'] = 'line'
- lines64_vector_n['Classe'] = 'line'
- hexagons64_vector_s['Classe'] = 'hexagon'
- hexagons64_vector_n['Classe'] = 'hexagon'
- ellipsis64_vector_s['Classe'] = 'ellipse'
- ellipsis64_vector_n['Classe'] = 'ellipse'
- squares16_vector_s['Classe'] = 'square'
- squares16_vector_n['Classe'] = 'square'
- circles16_vector_s['Classe'] = 'circle'
- circles16_vector_n['Classe'] = 'circle'
- triangles16_vector_s['Classe'] = 'triangle'
- triangles16_vector_n['Classe'] = 'triangle'
- trapezia16_vector_s['Classe'] = 'trapezia'
- trapezia16_vector_n['Classe'] = 'trapezia'
- rhombuses16_vector_s['Classe'] = 'rhombuse'
- rhombuses16_vector_n['Classe'] = 'rhombuse'
- rectangles16_vector_s['Classe'] = 'rectangle'
- rectangles16_vector_n['Classe'] = 'rectangle'
- lines16_vector_s['Classe'] = 'line'
- lines16_vector_n['Classe'] = 'line'
- hexagons16_vector_s['Classe'] = 'hexagon'
- hexagons16_vector_n['Classe'] = 'hexagon'
- ellipsis16_vector_s['Classe'] = 'ellipse'
- ellipsis16_vector_n['Classe'] = 'ellipse'
- dfs64_s = [squares64_vector_s,circles64_vector_s,triangles64_vector_s,trapezia64_vector_s,rhombuses64_vector_s,
- rectangles64_vector_s,lines64_vector_s,hexagons64_vector_s,ellipsis64_vector_s]
- dfs64_n = [squares64_vector_n,circles64_vector_n,triangles64_vector_n,trapezia64_vector_n,rhombuses64_vector_n,
- rectangles64_vector_n,lines64_vector_n,hexagons64_vector_n,ellipsis64_vector_n]
- dfs128_s = [squares128_vector_s,circles128_vector_s,triangles128_vector_s,trapezia128_vector_s,rhombuses128_vector_s,
- rectangles128_vector_s,lines128_vector_s,hexagons128_vector_s,ellipsis128_vector_s]
- dfs128_n = [squares128_vector_n,circles128_vector_n,triangles128_vector_n,trapezia128_vector_n,rhombuses128_vector_n,
- rectangles128_vector_n,lines128_vector_n,hexagons128_vector_n,ellipsis128_vector_n]
- dfs32_s = [squares32_vector_s,circles32_vector_s,triangles32_vector_s,trapezia32_vector_s,rhombuses32_vector_s,
- rectangles32_vector_s,lines32_vector_s,hexagons32_vector_s,ellipsis32_vector_s]
- dfs32_n = [squares32_vector_n,circles32_vector_n,triangles32_vector_n,trapezia32_vector_n,rhombuses32_vector_n,
- rectangles32_vector_n,lines32_vector_n,hexagons32_vector_n,ellipsis32_vector_n]
- dfs16_s = [squares16_vector_s,circles16_vector_s,triangles16_vector_s,trapezia16_vector_s,rhombuses16_vector_s,
- rectangles16_vector_s,lines16_vector_s,hexagons16_vector_s,ellipsis16_vector_s]
- dfs16_n = [squares16_vector_n,circles16_vector_n,triangles16_vector_n,trapezia16_vector_n,rhombuses16_vector_n,
- rectangles16_vector_n,lines16_vector_n,hexagons16_vector_n,ellipsis16_vector_n]
- # USANDO AS IMAGENS 128x128
- dataFrame128_s = pd.concat(dfs128_s, ignore_index=True)
- dataFrame128_2_s = dataFrame128_s.copy()
- del dataFrame128_s['Classe']
- dataFrame128_s.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_128_s_NOCLASS.csv')
- dataFrame128_s = fun.normalizar(dataFrame128_s)
- dataFrame128_s.fillna(0)
- dataFrame128_s['Classe'] = dataFrame128_2_s['Classe']
- dataFrame128_s.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_128_s.csv')
- dataFrame128_n = pd.concat(dfs128_n, ignore_index=True)
- dataFrame128_n.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_128_n.csv')
- dataFrame128_2_n = dataFrame128_n.copy()
- del dataFrame128_n['Classe']
- dataFrame128_n = fun.normalizar(dataFrame128_n)
- dataFrame128_n.fillna(0)
- dataFrame128_n['Classe'] = dataFrame128_2_n['Classe']
- dataFrame64_s = pd.concat(dfs64_s, ignore_index=True)
- print(dataFrame64_s)
- dataFrame64_s = dataFrame64_s.fillna(0)
- print(dataFrame64_s)
- dataFrame64_2_s = dataFrame64_s.copy()
- del dataFrame64_s['Classe']
- dataFrame64_s.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_64_sNOCLASS.csv')
- dataFrame64_s = fun.normalizar(dataFrame64_s)
- dataFrame64_s['Classe'] = dataFrame64_2_s['Classe']
- dataFrame64_s.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_64_s.csv')
- dataFrame64_n = pd.concat(dfs64_n, ignore_index=True)
- print(dataFrame64_n)
- dataFrame64_n.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_64_n.csv')
- dataFrame64_n = dataFrame64_n.fillna(0)
- print(dataFrame64_n)
- dataFrame64_2_n = dataFrame64_n.copy()
- del dataFrame64_n['Classe']
- dataFrame64_n = fun.normalizar(dataFrame64_n)
- dataFrame64_n['Classe'] = dataFrame64_2_n['Classe']
- dataFrame32_s = pd.concat(dfs32_s, ignore_index=True)
- dataFrame32_s = dataFrame32_s.fillna(0)
- dataFrame32_2_s = dataFrame32_s.copy()
- del dataFrame32_s['Classe']
- dataFrame32_s.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_32_sNOCLASS.csv')
- dataFrame32_s = fun.normalizar(dataFrame32_s)
- dataFrame32_s['Classe'] = dataFrame32_2_s['Classe']
- dataFrame32_s.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_32_s.csv')
- dataFrame32_n = pd.concat(dfs32_n, ignore_index=True)
- dataFrame32_n.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_32_n.csv')
- dataFrame32_n = dataFrame32_n.fillna(0)
- dataFrame32_2_n = dataFrame32_n.copy()
- del dataFrame32_n['Classe']
- dataFrame32_n = fun.normalizar(dataFrame32_n)
- dataFrame32_n['Classe'] = dataFrame32_2_n['Classe']
- dataFrame16_s = pd.concat(dfs16_s, ignore_index=True)
- dataFrame16_s = dataFrame16_s.fillna(0)
- dataFrame16_2_s = dataFrame16_s.copy()
- del dataFrame16_s['Classe']
- dataFrame16_s.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_16_sNOCLASS.csv')
- dataFrame16_s = fun.normalizar(dataFrame16_s)
- dataFrame16_s['Classe'] = dataFrame16_2_s['Classe']
- dataFrame16_s.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_16_s.csv')
- dataFrame16_n = pd.concat(dfs16_n, ignore_index=True)
- dataFrame16_n.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Images_16_n.csv')
- dataFrame16_n = dataFrame16_n.fillna(0)
- dataFrame16_2_n = dataFrame16_n.copy()
- del dataFrame16_n['Classe']
- dataFrame16_n = fun.normalizar(dataFrame16_n)
- dataFrame16_n['Classe'] = dataFrame16_2_n['Classe']
- # Criando o objeto do tipo k-folds com 10 folds
- # kfold = KFold(10, True, 1)
- # Criando o k-fold com 5 folds para execução do algoritmo genético
- kfold = KFold(5, True, 1)
- #Inicializando o Classificador do algoritmo genético
- # Random Forest Classifier
- RandomForest = RandomForestClassifier()
- RandomForest_acerto = []
- # RandomForest_tempo = []
- # RandomForest_precision = []
- # RandomForest_recall = []
- # RandomForest_fscore = []
- #criando a população com 20 cromossomos de tamanho 38
- cromossomos = fun.create_population(20, 38)
- for cromo in cromossomos:
- positions = fun.positions_chromossome(cromo)
- imagens, df_classe = fun.decode_chromossome(cromo)
- imagens = np.array(imagens)
- caracteristicas = fun.carac_imagens(positions, imagens)
- for x in range(0, 10):
- tempo1 = time.time()
- cols = list(df_classe.columns)
- cols.remove('Classe')
- df_images_noclass = df_classe[cols]
- df_images_class = df_classe['Classe']
- c = kfold.split(df_classe)
- for train_index, test_index in c:
- noclass_train, noclass_test = df_images_noclass.iloc[train_index], df_images_noclass.iloc[test_index]
- class_train, class_test = df_images_class.iloc[train_index], df_images_class.iloc[test_index]
- RandomForest_inicio = time.time()
- RandomForest.fit(noclass_train, class_train)
- RandomForest_acerto.append(RandomForest.score(noclass_test, class_test))
- df_classe = df_classe.sample(frac=1)
- print("Terminou a ", x)
- tempo2 = time.time()
- print("Tempo da rodada ", x, (tempo2 - tempo1) / 60)
- tempofinal = time.time()
- print('acuracia media ',np.mean(RandomForest_acerto))
- print('acuracia' )
- print(RandomForest_acerto)
- # Instanciando os algoritmos e seus vetores de tempo e acurácia
- '''
- #instanciando DataFrame com dados de saida
- DadosSaida = []
- DadosSaida = pd.DataFrame(DadosSaida)
- # Random Forest Classifier
- RandomForest = RandomForestClassifier()
- RandomForest_acerto = []
- RandomForest_tempo = []
- RandomForest_precision = []
- RandomForest_recall = []
- RandomForest_fscore = []
- # SVM com função de kernel linear
- SVMachine_L = SVC(kernel='linear')
- SVMachine_L_acerto = []
- SVMachine_L_tempo = []
- SVMachine_L_precision = []
- SVMachine_L_recall = []
- SVMachine_L_fscore = []
- #SVM com função de kernel RBF
- SVMachine_RBF = SVC(kernel='rbf', gamma='scale')
- SVMachine_RBF_acerto = []
- SVMachine_RBF_tempo = []
- SVMachine_RBF_precision = []
- SVMachine_RBF_recall = []
- SVMachine_RBF_fscore = []
- # KNN com k = 3, 5, 7
- KNN_3 = KNeighborsClassifier(n_neighbors=3, metric='euclidean')
- KNN_3_acerto = []
- KNN_3_tempo = []
- KNN_3_precision = []
- KNN_3_recall = []
- KNN_3_fscore = []
- KNN_5 = KNeighborsClassifier(n_neighbors=5, metric='euclidean')
- KNN_5_acerto = []
- KNN_5_tempo = []
- KNN_5_precision = []
- KNN_5_recall = []
- KNN_5_fscore = []
- KNN_7 = KNeighborsClassifier(n_neighbors=7, metric='euclidean')
- KNN_7_acerto = []
- KNN_7_tempo = []
- KNN_7_precision = []
- KNN_7_recall = []
- KNN_7_fscore = []
- # KNN Ponderado com k = 3, 5, 7
- KNNP_3 = KNeighborsClassifier(n_neighbors=3, weights='distance',metric='euclidean')
- KNNP_3_acerto = []
- KNNP_3_tempo = []
- KNNP_3_precision = []
- KNNP_3_recall = []
- KNNP_3_fscore = []
- KNNP_5 = KNeighborsClassifier(n_neighbors=5, weights='distance', metric='euclidean')
- KNNP_5_acerto = []
- KNNP_5_tempo = []
- KNNP_5_precision = []
- KNNP_5_recall = []
- KNNP_5_fscore = []
- KNNP_7 = KNeighborsClassifier(n_neighbors=7, weights='distance', metric='euclidean')
- KNNP_7_acerto = []
- KNNP_7_tempo = []
- KNNP_7_precision = []
- KNNP_7_recall = []
- KNNP_7_fscore = []
- # Naïve Bayes
- NaiveBayes = GaussianNB()
- NaiveBayes_acerto = []
- NaiveBayes_tempo = []
- NaiveBayes_precision = []
- NaiveBayes_recall = []
- NaiveBayes_fscore = []
- # Árvore de decisão
- DecisionTree = DecisionTreeClassifier()
- DecisionTree_acerto = []
- DecisionTree_tempo = []
- DecisionTree_precision = []
- DecisionTree_recall = []
- DecisionTree_fscore = []
- # MultiLayer Perceptron
- MLP = MLPClassifier()
- MLP_acerto = []
- MLP_tempo = []
- MLPP_precision = []
- MLP_recall = []
- MLP_fscore = []
- # Regressão Logística
- RegrLogistica = LogisticRegression(solver='lbfgs')
- RegrLogistica_acerto = []
- RegrLogistica_tempo = []
- RegreLogistica_precision = []
- RegrLogistica_recall = []
- RegrLogistica_fscore = []
- # ____________________ USANDO IMAGENS 128x128
- print('comecou o K fold')
- tempoinicial = time.time()
- for x in range(0, 10):
- tempo1 = time.time()
- cols = list(dataFrame128.columns)
- cols.remove('Classe')
- df_images_noclass = dataFrame128[cols]
- df_images_class = dataFrame128['Classe']
- c = kfold.split(dataFrame128)
- for train_index, test_index in c:
- noclass_train, noclass_test = df_images_noclass.iloc[train_index], df_images_noclass.iloc[test_index]
- class_train, class_test = df_images_class.iloc[train_index], df_images_class.iloc[test_index]
- KNN3_inicio = time.time()
- KNN_3.fit(noclass_train, class_train)
- KNN_3_acerto.append(KNN_3.score(noclass_test, class_test))
- KNN_3_recall. append(recall_score(class_test, KNN_3.predict(noclass_test),average='weighted'))
- KNN_3_precision.append(precision_score(class_test,KNN_3.predict(noclass_test),average='weighted'))
- KNN_3_fscore.append(f1_score(class_test,KNN_3.predict(noclass_test),average='weighted'))
- KNN3_fim = time.time()
- KNN_3_tempo.append(KNN3_fim - KNN3_inicio)
- KNN5_inicio = time.time()
- KNN_5.fit(noclass_train, class_train)
- KNN_5_acerto.append(KNN_5.score(noclass_test, class_test))
- KNN_5_recall.append(recall_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN_5_precision.append(precision_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN_5_fscore.append(f1_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN5_fim = time.time()
- KNN_5_tempo.append(KNN5_fim - KNN5_inicio)
- KNN7_inicio = time.time()
- KNN_7.fit(noclass_train, class_train)
- KNN_7_acerto.append(KNN_3.score(noclass_test, class_test))
- KNN_7_precision.append(precision_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN_7_recall.append(recall_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN_7_fscore.append(f1_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN7_fim = time.time()
- KNN_7_tempo.append(KNN3_fim - KNN3_inicio)
- KNNP3_inicio = time.time()
- KNNP_3.fit(noclass_train, class_train)
- KNNP_3_acerto.append(KNNP_3.score(noclass_test, class_test))
- KNNP_3_precision.append(precision_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP_3_recall.append(recall_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP_3_fscore.append(f1_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP3_fim = time.time()
- KNNP_3_tempo.append(KNNP3_fim - KNNP3_inicio)
- KNNP5_inicio = time.time()
- KNNP_5.fit(noclass_train, class_train)
- KNNP_5_acerto.append(KNNP_5.score(noclass_test, class_test))
- KNNP_5_precision.append(precision_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP_5_recall.append(recall_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP_5_fscore.append(f1_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP5_fim = time.time()
- KNNP_5_tempo.append(KNNP5_fim - KNNP5_inicio)
- KNNP7_inicio = time.time()
- KNNP_7.fit(noclass_train, class_train)
- KNNP_7_acerto.append(KNNP_7.score(noclass_test, class_test))
- KNNP_7_precision.append(precision_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP_7_recall.append(recall_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP_7_fscore.append(f1_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP7_fim = time.time()
- KNNP_7_tempo.append(KNNP7_fim - KNNP7_inicio)
- NaiveBayes_inicio = time.time()
- NaiveBayes.fit(noclass_train, class_train)
- NaiveBayes_acerto.append(NaiveBayes.score(noclass_test, class_test))
- NaiveBayes_precision.append(precision_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_recall.append(recall_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_fscore.append(f1_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_fim = time.time()
- NaiveBayes_tempo.append(NaiveBayes_fim - NaiveBayes_inicio)
- DecisionTree_inicio = time.time()
- DecisionTree.fit(noclass_train, class_train)
- DecisionTree_acerto.append(DecisionTree.score(noclass_test, class_test))
- DecisionTree_precision.append(precision_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_recall.append(recall_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_fscore.append(f1_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_fim = time.time()
- DecisionTree_tempo.append(DecisionTree_fim - DecisionTree_inicio)
- SVMachine_L_inicio = time.time()
- SVMachine_L.fit(noclass_train, class_train)
- SVMachine_L_acerto.append(SVMachine_L.score(noclass_test, class_test))
- SVMachine_L_precision.append(precision_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_recall.append(recall_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_fscore.append(f1_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_fim = time.time()
- SVMachine_L_tempo.append(SVMachine_L_fim - SVMachine_L_inicio)
- SVMachine_RBF_inicio = time.time()
- SVMachine_RBF.fit(noclass_train, class_train)
- SVMachine_RBF_acerto.append(SVMachine_RBF.score(noclass_test, class_test))
- SVMachine_RBF_recall.append(recall_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_precision.append(precision_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_fscore.append(f1_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_fim = time.time()
- SVMachine_RBF_tempo.append(SVMachine_RBF_fim - SVMachine_RBF_inicio)
- RegrLogistica_inicio = time.time()
- RegrLogistica.fit(noclass_train, class_train)
- RegrLogistica_acerto.append(RegrLogistica.score(noclass_test, class_test))
- RegreLogistica_precision.append(precision_score(class_test, (RegrLogistica.predict(noclass_test)),average='weighted'))
- RegrLogistica_recall.append(recall_score(class_test,RegrLogistica.predict(noclass_test),average='weighted'))
- RegrLogistica_fscore.append(f1_score(class_test,RegrLogistica.predict(noclass_test),average='weighted'))
- RegrLogistica_fim = time.time()
- RegrLogistica_tempo.append(RegrLogistica_fim - RegrLogistica_inicio)
- MLP_inicio = time.time()
- MLP.fit(noclass_train, class_train)
- MLP_acerto.append(MLP.score(noclass_test, class_test))
- MLPP_precision.append(precision_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_recall.append(recall_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_fscore.append(f1_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_fim = time.time()
- MLP_tempo.append(MLP_fim - MLP_inicio)
- RandomForest_inicio = time.time()
- RandomForest.fit(noclass_train, class_train)
- RandomForest_acerto.append(RandomForest.score(noclass_test, class_test))
- RandomForest_recall.append(recall_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_precision.append(precision_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_fscore.append(f1_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_fim = time.time()
- RandomForest_tempo.append(RandomForest_fim - RandomForest_inicio)
- dataFrame128 = dataFrame128.sample(frac=1)
- print("Terminou a ", x)
- tempo2 = time.time()
- print("Tempo da rodada ", x, (tempo2 - tempo1) / 60)
- tempofinal = time.time()
- fun.tendencia_central('KNN k = 3', KNN_3_acerto, KNN_3_tempo)
- fun.tendencia_central('KNN k = 5', KNN_5_acerto, KNN_5_tempo)
- fun.tendencia_central('KNN k = 7', KNN_7_acerto, KNN_7_tempo)
- fun.tendencia_central('KNN Ponderado k = 3', KNNP_3_acerto, KNNP_3_tempo)
- fun.tendencia_central('KNN Ponderado k = 5', KNNP_5_acerto, KNNP_5_tempo)
- fun.tendencia_central('KNN Ponderado k = 7', KNNP_7_acerto, KNNP_7_tempo)
- fun.tendencia_central('Naïve Bayes', NaiveBayes_acerto, NaiveBayes_tempo)
- fun.tendencia_central('Árvore de decisão', DecisionTree_acerto, DecisionTree_tempo)
- fun.tendencia_central('MLP', MLP_acerto, MLP_tempo)
- fun.tendencia_central('Regressão Logística', RegrLogistica_acerto, RegrLogistica_tempo)
- fun.tendencia_central('SVM linear', SVMachine_L_acerto, SVMachine_L_tempo)
- fun.tendencia_central('SVM RBF', SVMachine_RBF_acerto, SVMachine_RBF_tempo)
- fun.tendencia_central('Random Forest', RandomForest_acerto,RandomForest_tempo)
- Acuracia128 = [KNN_3_acerto,KNN_5_acerto,KNN_7_acerto,KNNP_3_acerto,KNNP_5_acerto,KNNP_7_acerto,
- NaiveBayes_acerto,DecisionTree_acerto,MLP_acerto,RegrLogistica_acerto,SVMachine_L_acerto,
- SVMachine_RBF_acerto,RandomForest_acerto]
- Precision128 = [KNN_3_precision,KNN_5_precision,KNN_7_precision,KNNP_3_precision,KNNP_5_precision,KNNP_7_precision,
- NaiveBayes_precision,DecisionTree_precision,MLPP_precision,RegreLogistica_precision,SVMachine_L_precision,
- SVMachine_RBF_precision,RandomForest_precision]
- Recall128 = [KNN_3_recall,KNN_5_recall,KNN_7_recall,KNNP_3_recall,KNNP_5_recall,KNNP_7_recall,NaiveBayes_recall,
- DecisionTree_recall,MLP_recall,RegrLogistica_recall,SVMachine_L_recall,SVMachine_RBF_recall,RandomForest_recall]
- Fscore128 = [KNN_3_fscore,KNN_5_fscore,KNN_7_fscore,KNNP_3_fscore,KNN_5_fscore,KNN_7_fscore,NaiveBayes_fscore,DecisionTree_fscore,
- MLP_fscore,RegrLogistica_fscore,SVMachine_L_fscore, SVMachine_RBF_fscore,RandomForest_fscore]
- Acuracia128 = pd.DataFrame(Acuracia128)
- Precision128 = pd.DataFrame(Precision128)
- Recall128 = pd.DataFrame(Recall128)
- Fscore128 = pd.DataFrame(Fscore128)
- Acuracia128.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Acuracia128.csv')
- Precision128.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Precision128.csv')
- Recall128.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Recall128.csv')
- Fscore128.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Fscore128.csv')
- mediasacuracias = {
- "KNN Ponderado k = 3": np.mean(KNNP_3_acerto),
- "KNN Ponderado k = 5": np.mean(KNNP_5_acerto),
- "KNN Ponderado k = 7": np.mean(KNNP_7_acerto),
- "Naive Bayes": np.mean(NaiveBayes_acerto),
- "KNN k = 3": np.mean(KNN_3_acerto),
- "KNN k = 5": np.mean(KNN_5_acerto),
- "KNN k = 7": np.mean(KNN_7_acerto),
- "Decision Tree": np.mean(DecisionTree_acerto),
- "SVM Linear": np.mean(SVMachine_L_acerto),
- "SVM RBF": np.mean(SVMachine_RBF_acerto),
- "Regressao Logistica": np.mean(RegrLogistica_acerto),
- "MLP": np.mean(MLP_acerto),
- "Random Forest": np.mean(RandomForest_acerto)
- }
- mediasacuracias = sorted(mediasacuracias.items(),
- key=lambda x: x[1])
- print(mediasacuracias)
- print("Tempo total: ", (tempofinal - tempoinicial) / 60)
- ## USANDO A BASE COM IMAGENS DE 64x64
- print('comecou o K fold')
- tempoinicial = time.time()
- for x in range(0, 10):
- tempo1 = time.time()
- cols = list(dataFrame64.columns)
- cols.remove('Classe')
- df_images_noclass = dataFrame64[cols]
- df_images_class = dataFrame64['Classe']
- c = kfold.split(dataFrame64)
- for train_index, test_index in c:
- noclass_train, noclass_test = df_images_noclass.iloc[train_index], df_images_noclass.iloc[test_index]
- class_train, class_test = df_images_class.iloc[train_index], df_images_class.iloc[test_index]
- KNN3_inicio = time.time()
- KNN_3.fit(noclass_train, class_train)
- KNN_3_acerto.append(KNN_3.score(noclass_test, class_test))
- KNN_3_recall. append(recall_score(class_test, KNN_3.predict(noclass_test),average='weighted'))
- KNN_3_precision.append(precision_score(class_test,KNN_3.predict(noclass_test),average='weighted'))
- KNN_3_fscore.append(f1_score(class_test,KNN_3.predict(noclass_test),average='weighted'))
- KNN3_fim = time.time()
- KNN_3_tempo.append(KNN3_fim - KNN3_inicio)
- KNN5_inicio = time.time()
- KNN_5.fit(noclass_train, class_train)
- KNN_5_acerto.append(KNN_5.score(noclass_test, class_test))
- KNN_5_recall.append(recall_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN_5_precision.append(precision_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN_5_fscore.append(f1_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN5_fim = time.time()
- KNN_5_tempo.append(KNN5_fim - KNN5_inicio)
- KNN7_inicio = time.time()
- KNN_7.fit(noclass_train, class_train)
- KNN_7_acerto.append(KNN_3.score(noclass_test, class_test))
- KNN_7_precision.append(precision_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN_7_recall.append(recall_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN_7_fscore.append(f1_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN7_fim = time.time()
- KNN_7_tempo.append(KNN3_fim - KNN3_inicio)
- KNNP3_inicio = time.time()
- KNNP_3.fit(noclass_train, class_train)
- KNNP_3_acerto.append(KNNP_3.score(noclass_test, class_test))
- KNNP_3_precision.append(precision_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP_3_recall.append(recall_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP_3_fscore.append(f1_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP3_fim = time.time()
- KNNP_3_tempo.append(KNNP3_fim - KNNP3_inicio)
- KNNP5_inicio = time.time()
- KNNP_5.fit(noclass_train, class_train)
- KNNP_5_acerto.append(KNNP_5.score(noclass_test, class_test))
- KNNP_5_precision.append(precision_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP_5_recall.append(recall_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP_5_fscore.append(f1_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP5_fim = time.time()
- KNNP_5_tempo.append(KNNP5_fim - KNNP5_inicio)
- KNNP7_inicio = time.time()
- KNNP_7.fit(noclass_train, class_train)
- KNNP_7_acerto.append(KNNP_7.score(noclass_test, class_test))
- KNNP_7_precision.append(precision_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP_7_recall.append(recall_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP_7_fscore.append(f1_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP7_fim = time.time()
- KNNP_7_tempo.append(KNNP7_fim - KNNP7_inicio)
- NaiveBayes_inicio = time.time()
- NaiveBayes.fit(noclass_train, class_train)
- NaiveBayes_acerto.append(NaiveBayes.score(noclass_test, class_test))
- NaiveBayes_precision.append(precision_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_recall.append(recall_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_fscore.append(f1_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_fim = time.time()
- NaiveBayes_tempo.append(NaiveBayes_fim - NaiveBayes_inicio)
- DecisionTree_inicio = time.time()
- DecisionTree.fit(noclass_train, class_train)
- DecisionTree_acerto.append(DecisionTree.score(noclass_test, class_test))
- DecisionTree_precision.append(precision_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_recall.append(recall_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_fscore.append(f1_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_fim = time.time()
- DecisionTree_tempo.append(DecisionTree_fim - DecisionTree_inicio)
- SVMachine_L_inicio = time.time()
- SVMachine_L.fit(noclass_train, class_train)
- SVMachine_L_acerto.append(SVMachine_L.score(noclass_test, class_test))
- SVMachine_L_precision.append(precision_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_recall.append(recall_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_fscore.append(f1_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_fim = time.time()
- SVMachine_L_tempo.append(SVMachine_L_fim - SVMachine_L_inicio)
- SVMachine_RBF_inicio = time.time()
- SVMachine_RBF.fit(noclass_train, class_train)
- SVMachine_RBF_acerto.append(SVMachine_RBF.score(noclass_test, class_test))
- SVMachine_RBF_recall.append(recall_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_precision.append(precision_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_fscore.append(f1_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_fim = time.time()
- SVMachine_RBF_tempo.append(SVMachine_RBF_fim - SVMachine_RBF_inicio)
- RegrLogistica_inicio = time.time()
- RegrLogistica.fit(noclass_train, class_train)
- RegrLogistica_acerto.append(RegrLogistica.score(noclass_test, class_test))
- RegreLogistica_precision.append(precision_score(class_test, (RegrLogistica.predict(noclass_test)),average='weighted'))
- RegrLogistica_recall.append(recall_score(class_test,RegrLogistica.predict(noclass_test),average='weighted'))
- RegrLogistica_fscore.append(f1_score(class_test,RegrLogistica.predict(noclass_test),average='weighted'))
- RegrLogistica_fim = time.time()
- RegrLogistica_tempo.append(RegrLogistica_fim - RegrLogistica_inicio)
- MLP_inicio = time.time()
- MLP.fit(noclass_train, class_train)
- MLP_acerto.append(MLP.score(noclass_test, class_test))
- MLPP_precision.append(precision_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_recall.append(recall_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_fscore.append(f1_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_fim = time.time()
- MLP_tempo.append(MLP_fim - MLP_inicio)
- RandomForest_inicio = time.time()
- RandomForest.fit(noclass_train, class_train)
- RandomForest_acerto.append(RandomForest.score(noclass_test, class_test))
- RandomForest_recall.append(recall_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_precision.append(precision_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_fscore.append(f1_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_fim = time.time()
- RandomForest_tempo.append(RandomForest_fim - RandomForest_inicio)
- dataFrame64 = dataFrame64.sample(frac=1)
- print("Terminou a ", x)
- tempo2 = time.time()
- print("Tempo da rodada ", x, (tempo2 - tempo1) / 60)
- tempofinal = time.time()
- fun.tendencia_central('KNN k = 3', KNN_3_acerto, KNN_3_tempo)
- fun.tendencia_central('KNN k = 5', KNN_5_acerto, KNN_5_tempo)
- fun.tendencia_central('KNN k = 7', KNN_7_acerto, KNN_7_tempo)
- fun.tendencia_central('KNN Ponderado k = 3', KNNP_3_acerto, KNNP_3_tempo)
- fun.tendencia_central('KNN Ponderado k = 5', KNNP_5_acerto, KNNP_5_tempo)
- fun.tendencia_central('KNN Ponderado k = 7', KNNP_7_acerto, KNNP_7_tempo)
- fun.tendencia_central('Naïve Bayes', NaiveBayes_acerto, NaiveBayes_tempo)
- fun.tendencia_central('Árvore de decisão', DecisionTree_acerto, DecisionTree_tempo)
- fun.tendencia_central('MLP', MLP_acerto, MLP_tempo)
- fun.tendencia_central('Regressão Logística', RegrLogistica_acerto, RegrLogistica_tempo)
- fun.tendencia_central('SVM linear', SVMachine_L_acerto, SVMachine_L_tempo)
- fun.tendencia_central('SVM RBF', SVMachine_RBF_acerto, SVMachine_RBF_tempo)
- fun.tendencia_central('Random Forest', RandomForest_acerto,RandomForest_tempo)
- mediasacuracias = {
- "KNN Ponderado k = 3": np.mean(KNNP_3_acerto),
- "KNN Ponderado k = 5": np.mean(KNNP_5_acerto),
- "KNN Ponderado k = 7": np.mean(KNNP_7_acerto),
- "Naive Bayes": np.mean(NaiveBayes_acerto),
- "KNN k = 3": np.mean(KNN_3_acerto),
- "KNN k = 5": np.mean(KNN_5_acerto),
- "KNN k = 7": np.mean(KNN_7_acerto),
- "Decision Tree": np.mean(DecisionTree_acerto),
- "SVM Linear": np.mean(SVMachine_L_acerto),
- "SVM RBF": np.mean(SVMachine_RBF_acerto),
- "Regressao Logistica": np.mean(RegrLogistica_acerto),
- "MLP": np.mean(MLP_acerto),
- "Random Forest": np.mean(RandomForest_acerto)
- }
- mediasacuracias = sorted(mediasacuracias.items(),
- key=lambda x: x[1])
- print(mediasacuracias)
- print("Tempo total: ", (tempofinal - tempoinicial) / 60)
- Acuracia64 = [KNN_3_acerto,KNN_5_acerto,KNN_7_acerto,KNNP_3_acerto,KNNP_5_acerto,KNNP_7_acerto,
- NaiveBayes_acerto,DecisionTree_acerto,MLP_acerto,RegrLogistica_acerto,SVMachine_L_acerto,
- SVMachine_RBF_acerto,RandomForest_acerto]
- Precision64 = [KNN_3_precision,KNN_5_precision,KNN_7_precision,KNNP_3_precision,KNNP_5_precision,KNNP_7_precision,
- NaiveBayes_precision,DecisionTree_precision,MLPP_precision,RegreLogistica_precision,SVMachine_L_precision,
- SVMachine_RBF_precision,RandomForest_precision]
- Recall64 = [KNN_3_recall,KNN_5_recall,KNN_7_recall,KNNP_3_recall,KNNP_5_recall,KNNP_7_recall,NaiveBayes_recall,
- DecisionTree_recall,MLP_recall,RegrLogistica_recall,SVMachine_L_recall,SVMachine_RBF_recall,RandomForest_recall]
- Fscore64 = [KNN_3_fscore,KNN_5_fscore,KNN_7_fscore,KNNP_3_fscore,KNN_5_fscore,KNN_7_fscore,NaiveBayes_fscore,DecisionTree_fscore,
- MLP_fscore,RegrLogistica_fscore,SVMachine_L_fscore, SVMachine_RBF_fscore,RandomForest_fscore]
- Acuracia64 = pd.DataFrame(Acuracia64)
- Precision64 = pd.DataFrame(Precision64)
- Recall64 = pd.DataFrame(Recall64)
- Fscore64 = pd.DataFrame(Fscore64)
- Acuracia64.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Acuracia64.csv')
- Precision64.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Precision64.csv')
- Recall64.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Recall64.csv')
- Fscore64.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Fscore64.csv')
- # USANDO BASE DE 32x32
- print('comecou o K fold')
- tempoinicial = time.time()
- for x in range(0, 10):
- tempo1 = time.time()
- cols = list(dataFrame32.columns)
- cols.remove('Classe')
- df_images_noclass = dataFrame32[cols]
- df_images_class = dataFrame32['Classe']
- c = kfold.split(dataFrame32)
- for train_index, test_index in c:
- noclass_train, noclass_test = df_images_noclass.iloc[train_index], df_images_noclass.iloc[test_index]
- class_train, class_test = df_images_class.iloc[train_index], df_images_class.iloc[test_index]
- KNN3_inicio = time.time()
- KNN_3.fit(noclass_train, class_train)
- KNN_3_acerto.append(KNN_3.score(noclass_test, class_test))
- KNN_3_recall. append(recall_score(class_test, KNN_3.predict(noclass_test),average='weighted'))
- KNN_3_precision.append(precision_score(class_test,KNN_3.predict(noclass_test),average='weighted'))
- KNN_3_fscore.append(f1_score(class_test,KNN_3.predict(noclass_test),average='weighted'))
- KNN3_fim = time.time()
- KNN_3_tempo.append(KNN3_fim - KNN3_inicio)
- KNN5_inicio = time.time()
- KNN_5.fit(noclass_train, class_train)
- KNN_5_acerto.append(KNN_5.score(noclass_test, class_test))
- KNN_5_recall.append(recall_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN_5_precision.append(precision_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN_5_fscore.append(f1_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN5_fim = time.time()
- KNN_5_tempo.append(KNN5_fim - KNN5_inicio)
- KNN7_inicio = time.time()
- KNN_7.fit(noclass_train, class_train)
- KNN_7_acerto.append(KNN_3.score(noclass_test, class_test))
- KNN_7_precision.append(precision_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN_7_recall.append(recall_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN_7_fscore.append(f1_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN7_fim = time.time()
- KNN_7_tempo.append(KNN3_fim - KNN3_inicio)
- KNNP3_inicio = time.time()
- KNNP_3.fit(noclass_train, class_train)
- KNNP_3_acerto.append(KNNP_3.score(noclass_test, class_test))
- KNNP_3_precision.append(precision_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP_3_recall.append(recall_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP_3_fscore.append(f1_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP3_fim = time.time()
- KNNP_3_tempo.append(KNNP3_fim - KNNP3_inicio)
- KNNP5_inicio = time.time()
- KNNP_5.fit(noclass_train, class_train)
- KNNP_5_acerto.append(KNNP_5.score(noclass_test, class_test))
- KNNP_5_precision.append(precision_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP_5_recall.append(recall_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP_5_fscore.append(f1_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP5_fim = time.time()
- KNNP_5_tempo.append(KNNP5_fim - KNNP5_inicio)
- KNNP7_inicio = time.time()
- KNNP_7.fit(noclass_train, class_train)
- KNNP_7_acerto.append(KNNP_7.score(noclass_test, class_test))
- KNNP_7_precision.append(precision_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP_7_recall.append(recall_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP_7_fscore.append(f1_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP7_fim = time.time()
- KNNP_7_tempo.append(KNNP7_fim - KNNP7_inicio)
- NaiveBayes_inicio = time.time()
- NaiveBayes.fit(noclass_train, class_train)
- NaiveBayes_acerto.append(NaiveBayes.score(noclass_test, class_test))
- NaiveBayes_precision.append(precision_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_recall.append(recall_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_fscore.append(f1_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_fim = time.time()
- NaiveBayes_tempo.append(NaiveBayes_fim - NaiveBayes_inicio)
- DecisionTree_inicio = time.time()
- DecisionTree.fit(noclass_train, class_train)
- DecisionTree_acerto.append(DecisionTree.score(noclass_test, class_test))
- DecisionTree_precision.append(precision_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_recall.append(recall_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_fscore.append(f1_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_fim = time.time()
- DecisionTree_tempo.append(DecisionTree_fim - DecisionTree_inicio)
- SVMachine_L_inicio = time.time()
- SVMachine_L.fit(noclass_train, class_train)
- SVMachine_L_acerto.append(SVMachine_L.score(noclass_test, class_test))
- SVMachine_L_precision.append(precision_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_recall.append(recall_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_fscore.append(f1_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_fim = time.time()
- SVMachine_L_tempo.append(SVMachine_L_fim - SVMachine_L_inicio)
- SVMachine_RBF_inicio = time.time()
- SVMachine_RBF.fit(noclass_train, class_train)
- SVMachine_RBF_acerto.append(SVMachine_RBF.score(noclass_test, class_test))
- SVMachine_RBF_recall.append(recall_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_precision.append(precision_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_fscore.append(f1_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_fim = time.time()
- SVMachine_RBF_tempo.append(SVMachine_RBF_fim - SVMachine_RBF_inicio)
- RegrLogistica_inicio = time.time()
- RegrLogistica.fit(noclass_train, class_train)
- RegrLogistica_acerto.append(RegrLogistica.score(noclass_test, class_test))
- RegreLogistica_precision.append(precision_score(class_test, (RegrLogistica.predict(noclass_test)),average='weighted'))
- RegrLogistica_recall.append(recall_score(class_test,RegrLogistica.predict(noclass_test),average='weighted'))
- RegrLogistica_fscore.append(f1_score(class_test,RegrLogistica.predict(noclass_test),average='weighted'))
- RegrLogistica_fim = time.time()
- RegrLogistica_tempo.append(RegrLogistica_fim - RegrLogistica_inicio)
- MLP_inicio = time.time()
- MLP.fit(noclass_train, class_train)
- MLP_acerto.append(MLP.score(noclass_test, class_test))
- MLPP_precision.append(precision_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_recall.append(recall_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_fscore.append(f1_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_fim = time.time()
- MLP_tempo.append(MLP_fim - MLP_inicio)
- RandomForest_inicio = time.time()
- RandomForest.fit(noclass_train, class_train)
- RandomForest_acerto.append(RandomForest.score(noclass_test, class_test))
- RandomForest_recall.append(recall_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_precision.append(precision_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_fscore.append(f1_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_fim = time.time()
- RandomForest_tempo.append(RandomForest_fim - RandomForest_inicio)
- dataFrame32 = dataFrame32.sample(frac=1)
- print("Terminou a ", x)
- tempo2 = time.time()
- print("Tempo da rodada ", x, (tempo2 - tempo1) / 60)
- tempofinal = time.time()
- fun.tendencia_central('KNN k = 3', KNN_3_acerto, KNN_3_tempo)
- fun.tendencia_central('KNN k = 5', KNN_5_acerto, KNN_5_tempo)
- fun.tendencia_central('KNN k = 7', KNN_7_acerto, KNN_7_tempo)
- fun.tendencia_central('KNN Ponderado k = 3', KNNP_3_acerto, KNNP_3_tempo)
- fun.tendencia_central('KNN Ponderado k = 5', KNNP_5_acerto, KNNP_5_tempo)
- fun.tendencia_central('KNN Ponderado k = 7', KNNP_7_acerto, KNNP_7_tempo)
- fun.tendencia_central('Naïve Bayes', NaiveBayes_acerto, NaiveBayes_tempo)
- fun.tendencia_central('Árvore de decisão', DecisionTree_acerto, DecisionTree_tempo)
- fun.tendencia_central('MLP', MLP_acerto, MLP_tempo)
- fun.tendencia_central('Regressão Logística', RegrLogistica_acerto, RegrLogistica_tempo)
- fun.tendencia_central('SVM linear', SVMachine_L_acerto, SVMachine_L_tempo)
- fun.tendencia_central('SVM RBF', SVMachine_RBF_acerto, SVMachine_RBF_tempo)
- fun.tendencia_central('Random Forest', RandomForest_acerto,RandomForest_tempo)
- mediasacuracias = {
- "KNN Ponderado k = 3": np.mean(KNNP_3_acerto),
- "KNN Ponderado k = 5": np.mean(KNNP_5_acerto),
- "KNN Ponderado k = 7": np.mean(KNNP_7_acerto),
- "Naive Bayes": np.mean(NaiveBayes_acerto),
- "KNN k = 3": np.mean(KNN_3_acerto),
- "KNN k = 5": np.mean(KNN_5_acerto),
- "KNN k = 7": np.mean(KNN_7_acerto),
- "Decision Tree": np.mean(DecisionTree_acerto),
- "SVM Linear": np.mean(SVMachine_L_acerto),
- "SVM RBF": np.mean(SVMachine_RBF_acerto),
- "Regressao Logistica": np.mean(RegrLogistica_acerto),
- "MLP": np.mean(MLP_acerto),
- "Random Forest": np.mean(RandomForest_acerto)
- }
- mediasacuracias = sorted(mediasacuracias.items(),
- key=lambda x: x[1])
- print(mediasacuracias)
- print("Tempo total: ", (tempofinal - tempoinicial) / 60)
- Acuracia32 = [KNN_3_acerto,KNN_5_acerto,KNN_7_acerto,KNNP_3_acerto,KNNP_5_acerto,KNNP_7_acerto,
- NaiveBayes_acerto,DecisionTree_acerto,MLP_acerto,RegrLogistica_acerto,SVMachine_L_acerto,
- SVMachine_RBF_acerto,RandomForest_acerto]
- Precision32 = [KNN_3_precision,KNN_5_precision,KNN_7_precision,KNNP_3_precision,KNNP_5_precision,KNNP_7_precision,
- NaiveBayes_precision,DecisionTree_precision,MLPP_precision,RegreLogistica_precision,SVMachine_L_precision,
- SVMachine_RBF_precision,RandomForest_precision]
- Recall32 = [KNN_3_recall,KNN_5_recall,KNN_7_recall,KNNP_3_recall,KNNP_5_recall,KNNP_7_recall,NaiveBayes_recall,
- DecisionTree_recall,MLP_recall,RegrLogistica_recall,SVMachine_L_recall,SVMachine_RBF_recall,RandomForest_recall]
- Fscore32 = [KNN_3_fscore,KNN_5_fscore,KNN_7_fscore,KNNP_3_fscore,KNN_5_fscore,KNN_7_fscore,NaiveBayes_fscore,DecisionTree_fscore,
- MLP_fscore,RegrLogistica_fscore,SVMachine_L_fscore, SVMachine_RBF_fscore,RandomForest_fscore]
- Acuracia32 = pd.DataFrame(Acuracia32)
- Precision32 = pd.DataFrame(Precision32)
- Recall32 = pd.DataFrame(Recall32)
- Fscore32 = pd.DataFrame(Fscore32)
- Acuracia32.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Acuracia32.csv')
- Precision32.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Precision32.csv')
- Recall32.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Recall32.csv')
- Fscore32.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Fscore32.csv')
- '''
- '''
- # USANDO A BASE COM IMAGENS 16x16
- print('comecou o K fold')
- tempoinicial = time.time()
- for x in range(0, 10):
- tempo1 = time.time()
- cols = list(dataFrame16.columns)
- cols.remove('Classe')
- df_images_noclass = dataFrame16[cols]
- df_images_class = dataFrame16['Classe']
- c = kfold.split(dataFrame16)
- for train_index, test_index in c:
- noclass_train, noclass_test = df_images_noclass.iloc[train_index], df_images_noclass.iloc[test_index]
- class_train, class_test = df_images_class.iloc[train_index], df_images_class.iloc[test_index]
- KNN3_inicio = time.time()
- KNN_3.fit(noclass_train, class_train)
- KNN_3_acerto.append(KNN_3.score(noclass_test, class_test))
- KNN_3_recall. append(recall_score(class_test, KNN_3.predict(noclass_test),average='weighted'))
- KNN_3_precision.append(precision_score(class_test,KNN_3.predict(noclass_test),average='weighted'))
- KNN_3_fscore.append(f1_score(class_test,KNN_3.predict(noclass_test),average='weighted'))
- KNN3_fim = time.time()
- KNN_3_tempo.append(KNN3_fim - KNN3_inicio)
- KNN5_inicio = time.time()
- KNN_5.fit(noclass_train, class_train)
- KNN_5_acerto.append(KNN_5.score(noclass_test, class_test))
- KNN_5_recall.append(recall_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN_5_precision.append(precision_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN_5_fscore.append(f1_score(class_test,KNN_5.predict(noclass_test),average='weighted'))
- KNN5_fim = time.time()
- KNN_5_tempo.append(KNN5_fim - KNN5_inicio)
- KNN7_inicio = time.time()
- KNN_7.fit(noclass_train, class_train)
- KNN_7_acerto.append(KNN_3.score(noclass_test, class_test))
- KNN_7_precision.append(precision_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN_7_recall.append(recall_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN_7_fscore.append(f1_score(class_test,KNN_7.predict(noclass_test),average='weighted'))
- KNN7_fim = time.time()
- KNN_7_tempo.append(KNN3_fim - KNN3_inicio)
- KNNP3_inicio = time.time()
- KNNP_3.fit(noclass_train, class_train)
- KNNP_3_acerto.append(KNNP_3.score(noclass_test, class_test))
- KNNP_3_precision.append(precision_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP_3_recall.append(recall_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP_3_fscore.append(f1_score(class_test,KNNP_3.predict(noclass_test),average='weighted'))
- KNNP3_fim = time.time()
- KNNP_3_tempo.append(KNNP3_fim - KNNP3_inicio)
- KNNP5_inicio = time.time()
- KNNP_5.fit(noclass_train, class_train)
- KNNP_5_acerto.append(KNNP_5.score(noclass_test, class_test))
- KNNP_5_precision.append(precision_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP_5_recall.append(recall_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP_5_fscore.append(f1_score(class_test,KNNP_5.predict(noclass_test),average='weighted'))
- KNNP5_fim = time.time()
- KNNP_5_tempo.append(KNNP5_fim - KNNP5_inicio)
- KNNP7_inicio = time.time()
- KNNP_7.fit(noclass_train, class_train)
- KNNP_7_acerto.append(KNNP_7.score(noclass_test, class_test))
- KNNP_7_precision.append(precision_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP_7_recall.append(recall_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP_7_fscore.append(f1_score(class_test,KNNP_7.predict(noclass_test),average='weighted'))
- KNNP7_fim = time.time()
- KNNP_7_tempo.append(KNNP7_fim - KNNP7_inicio)
- NaiveBayes_inicio = time.time()
- NaiveBayes.fit(noclass_train, class_train)
- NaiveBayes_acerto.append(NaiveBayes.score(noclass_test, class_test))
- NaiveBayes_precision.append(precision_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_recall.append(recall_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_fscore.append(f1_score(class_test,NaiveBayes.predict(noclass_test),average='weighted'))
- NaiveBayes_fim = time.time()
- NaiveBayes_tempo.append(NaiveBayes_fim - NaiveBayes_inicio)
- DecisionTree_inicio = time.time()
- DecisionTree.fit(noclass_train, class_train)
- DecisionTree_acerto.append(DecisionTree.score(noclass_test, class_test))
- DecisionTree_precision.append(precision_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_recall.append(recall_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_fscore.append(f1_score(class_test,DecisionTree.predict(noclass_test),average='weighted'))
- DecisionTree_fim = time.time()
- DecisionTree_tempo.append(DecisionTree_fim - DecisionTree_inicio)
- SVMachine_L_inicio = time.time()
- SVMachine_L.fit(noclass_train, class_train)
- SVMachine_L_acerto.append(SVMachine_L.score(noclass_test, class_test))
- SVMachine_L_precision.append(precision_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_recall.append(recall_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_fscore.append(f1_score(class_test,SVMachine_L.predict(noclass_test),average='weighted'))
- SVMachine_L_fim = time.time()
- SVMachine_L_tempo.append(SVMachine_L_fim - SVMachine_L_inicio)
- SVMachine_RBF_inicio = time.time()
- SVMachine_RBF.fit(noclass_train, class_train)
- SVMachine_RBF_acerto.append(SVMachine_RBF.score(noclass_test, class_test))
- SVMachine_RBF_recall.append(recall_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_precision.append(precision_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_fscore.append(f1_score(class_test,SVMachine_RBF.predict(noclass_test),average='weighted'))
- SVMachine_RBF_fim = time.time()
- SVMachine_RBF_tempo.append(SVMachine_RBF_fim - SVMachine_RBF_inicio)
- RegrLogistica_inicio = time.time()
- RegrLogistica.fit(noclass_train, class_train)
- RegrLogistica_acerto.append(RegrLogistica.score(noclass_test, class_test))
- RegreLogistica_precision.append(precision_score(class_test, (RegrLogistica.predict(noclass_test)),average='weighted'))
- RegrLogistica_recall.append(recall_score(class_test,RegrLogistica.predict(noclass_test),average='weighted'))
- RegrLogistica_fscore.append(f1_score(class_test,RegrLogistica.predict(noclass_test),average='weighted'))
- RegrLogistica_fim = time.time()
- RegrLogistica_tempo.append(RegrLogistica_fim - RegrLogistica_inicio)
- MLP_inicio = time.time()
- MLP.fit(noclass_train, class_train)
- MLP_acerto.append(MLP.score(noclass_test, class_test))
- MLPP_precision.append(precision_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_recall.append(recall_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_fscore.append(f1_score(class_test,MLP.predict(noclass_test),average='weighted'))
- MLP_fim = time.time()
- MLP_tempo.append(MLP_fim - MLP_inicio)
- RandomForest_inicio = time.time()
- RandomForest.fit(noclass_train, class_train)
- RandomForest_acerto.append(RandomForest.score(noclass_test, class_test))
- RandomForest_recall.append(recall_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_precision.append(precision_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_fscore.append(f1_score(class_test,RandomForest.predict(noclass_test),average='weighted'))
- RandomForest_fim = time.time()
- RandomForest_tempo.append(RandomForest_fim - RandomForest_inicio)
- dataFrame16 = dataFrame16.sample(frac=1)
- print("Terminou a ", x)
- tempo2 = time.time()
- print("Tempo da rodada ", x, (tempo2 - tempo1) / 60)
- tempofinal = time.time()
- fun.tendencia_central('KNN k = 3', KNN_3_acerto, KNN_3_tempo)
- fun.tendencia_central('KNN k = 5', KNN_5_acerto, KNN_5_tempo)
- fun.tendencia_central('KNN k = 7', KNN_7_acerto, KNN_7_tempo)
- fun.tendencia_central('KNN Ponderado k = 3', KNNP_3_acerto, KNNP_3_tempo)
- fun.tendencia_central('KNN Ponderado k = 5', KNNP_5_acerto, KNNP_5_tempo)
- fun.tendencia_central('KNN Ponderado k = 7', KNNP_7_acerto, KNNP_7_tempo)
- fun.tendencia_central('Naïve Bayes', NaiveBayes_acerto, NaiveBayes_tempo)
- fun.tendencia_central('Árvore de decisão', DecisionTree_acerto, DecisionTree_tempo)
- fun.tendencia_central('MLP', MLP_acerto, MLP_tempo)
- fun.tendencia_central('Regressão Logística', RegrLogistica_acerto, RegrLogistica_tempo)
- fun.tendencia_central('SVM linear', SVMachine_L_acerto, SVMachine_L_tempo)
- fun.tendencia_central('SVM RBF', SVMachine_RBF_acerto, SVMachine_RBF_tempo)
- fun.tendencia_central('Random Forest', RandomForest_acerto,RandomForest_tempo)
- mediasacuracias = {
- "KNN Ponderado k = 3": np.mean(KNNP_3_acerto),
- "KNN Ponderado k = 5": np.mean(KNNP_5_acerto),
- "KNN Ponderado k = 7": np.mean(KNNP_7_acerto),
- "Naive Bayes": np.mean(NaiveBayes_acerto),
- "KNN k = 3": np.mean(KNN_3_acerto),
- "KNN k = 5": np.mean(KNN_5_acerto),
- "KNN k = 7": np.mean(KNN_7_acerto),
- "Decision Tree": np.mean(DecisionTree_acerto),
- "SVM Linear": np.mean(SVMachine_L_acerto),
- "SVM RBF": np.mean(SVMachine_RBF_acerto),
- "Regressao Logistica": np.mean(RegrLogistica_acerto),
- "MLP": np.mean(MLP_acerto),
- "Random Forest": np.mean(RandomForest_acerto)
- }
- mediasacuracias = sorted(mediasacuracias.items(),
- key=lambda x: x[1])
- print(mediasacuracias)
- print("Tempo total: ", (tempofinal - tempoinicial) / 60)
- Acuracia16 = [KNN_3_acerto,KNN_5_acerto,KNN_7_acerto,KNNP_3_acerto,KNNP_5_acerto,KNNP_7_acerto,
- NaiveBayes_acerto,DecisionTree_acerto,MLP_acerto,RegrLogistica_acerto,SVMachine_L_acerto,
- SVMachine_RBF_acerto,RandomForest_acerto]
- Precision16 = [KNN_3_precision,KNN_5_precision,KNN_7_precision,KNNP_3_precision,KNNP_5_precision,KNNP_7_precision,
- NaiveBayes_precision,DecisionTree_precision,MLPP_precision,RegreLogistica_precision,SVMachine_L_precision,
- SVMachine_RBF_precision,RandomForest_precision]
- Recall16 = [KNN_3_recall,KNN_5_recall,KNN_7_recall,KNNP_3_recall,KNNP_5_recall,KNNP_7_recall,NaiveBayes_recall,
- DecisionTree_recall,MLP_recall,RegrLogistica_recall,SVMachine_L_recall,SVMachine_RBF_recall,RandomForest_recall]
- Fscore16 = [KNN_3_fscore,KNN_5_fscore,KNN_7_fscore,KNNP_3_fscore,KNN_5_fscore,KNN_7_fscore,NaiveBayes_fscore,DecisionTree_fscore,
- MLP_fscore,RegrLogistica_fscore,SVMachine_L_fscore, SVMachine_RBF_fscore,RandomForest_fscore]
- Acuracia16 = pd.DataFrame(Acuracia16)
- Precision16 = pd.DataFrame(Precision16)
- Recall16 = pd.DataFrame(Recall16)
- Fscore16 = pd.DataFrame(Fscore16)
- Acuracia16.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Acuracia16.csv')
- Precision16.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Precision16.csv')
- Recall16.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Recall16.csv')
- Fscore16.to_csv('C:/Users/Auricelia/Desktop/DataSetsML/Fscore16.csv')
- '''
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement