机器学习KNN算法分类kaggle猫狗

KNN算法思想就不多说了，很简单，基于欧氏距离或者Mahantta距离进行计算，然后投票决定你的类别。

这里用三个文件;

simplepreprocessor.py

import cv2

class SimplePreprocessor:
	def __init__(self, width, height, inter=cv2.INTER_AREA):
		self.width = width
		self.height = height
		self.inter = inter
	# 设置输入进来的图片统一到一个尺寸
	def preprocess(self, image):
		return cv2.resize(image, (self.width, self.height), interpolation=self.inter)


if __name__ == '__main__':

	s = SimplePreprocessor(300, 400)
	img = cv2.imread('beauty.jpg')
	# print(img)
	cv2.imshow('src', img)
	cv2.imshow("resize", s.preprocess(img))

	cv2.waitKey(0)
	# cv2.destroyallWindows()

simpledatasetloader.py

import numpy as np
import cv2
import os
'''
class SimplePreprocessor:
	def __init__(self, width, height, inter=cv2.INTER_AREA):
		self.width = width
		self.height = height
		self.inter = inter
	@staticmethod
	def preprocess(self, image):
		return cv2.resize(image, (self.width, self.height), interpolation=self.inter)
'''

class SimpleDatasetLoader:
	def __init__(self, preprocessors=None):
		self.preprocessors = preprocessors

		if self.preprocessors is None:
			self.preprocessors = []

	def load(self, imagePaths, verbose=-1):
		'''
		data_cat = []
		data_dog = []
		labels_dog = []
		labels_cat = []
		'''
		# 接收图像数据(像素值)和相对应的label
		data = []
		labels = []

		for (i, imagePath) in enumerate(imagePaths):
			image = cv2.imread(imagePath)
			# 根据文件夹的分类进行获取label
			# 文件夹应该对应
			# dog_...  cat_...
			label = imagePath.split(os.path.sep)[-2]
			if self.preprocessors is not None:
				for p in self.preprocessors:
					image = p.preprocess(image)
			data.append(image)
			labels.append(label)
			# 每处理500张就输出信息
			if verbose > 0 and i > 0 and (i + 1)%verbose == 0:
				print('[INFO] processed {}/{}'.format(i+1, len(imagePaths)))

		# print(data_cat)
		# print(data_dog)
		# print(labels_cat)
		# print(labels_dog)

		return (np.array(data), np.array(labels))


if __name__ == '__main__':
	s = SimpleDatasetLoader()
	imagePaths = '/home/king/test/python/train/data/all/train'
	s.load(imagePaths)

knn.py

from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report
from simplepreprocessor import SimplePreprocessor
from simpledatasetloader import SimpleDatasetLoader
from imutils import paths
import argparse

if __name__ == '__main__':
	# 命令行参数设置
	ap = argparse.ArgumentParser()
	ap.add_argument("-d", "--dataset", required=True, help="path to input dataset")
	ap.add_argument("-k", "--neighbors", type=int, default=1, help="of nearest neighbors for classification")
	ap.add_argument("-j", "--jobs", type=int, help="of jobs for K-NN distance (-1 uses all variables cores)")
	args = vars(ap.parse_args())

	print("[INFO] loading images...")
	# 加载数据集的文件路径
	imagePaths = list(paths.list_images(args["dataset"]))
	# 对数据集文件夹下的图片进行预处理，统一到32x32的尺寸
	sp = SimplePreprocessor(32, 32)
	sdl = SimpleDatasetLoader(preprocessors=[sp])
	# 从RGB三颜色通道flat到1维矩阵
	(data, labels) = sdl.load(imagePaths, verbose=500)
	data = data.reshape((data.shape[0], 3072))

	print("[INFO] features matrix:{:.1f}MB".format(data.nbytes / (1024*1000.0)))
	# 对类别进行编码，比如dog-0,cat-1
	le = LabelEncoder()
	labels = le.fit_transform(labels)
	# print(labels)
	# 训练集70%用来训练，25%用来测试，其实也是相当于validation
	(trainX, testX, trainY, testY) = train_test_split(data, labels, test_size=0.25, random_state=42)

	# print(trainX)
	# print(trainY)

	print("[INFO] evaluating K-NN classifier...")
	# 构建KNN分类器
	model = KNeighborsClassifier(n_neighbors=args["neighbors"], n_jobs=args["jobs"])
	# 模型训练
	model.fit(trainX, trainY)
	print(classification_report(testY, model.predict(testX), target_names=le.classes_))