机器学习逻辑回归分类，图片 0-9 分类

""" 
@author:Jerry Liu
@file: DigitsLogisticRegression.py 
@version:
@time: 2018/12/11 
@email:liudongbing8@163.com
@function： https://www.jianshu.com/p/0e6fb477a159
"""

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import load_digits
from sklearn import metrics
# 划分训练与测试数据集
from sklearn.model_selection import train_test_split

# Logistic Regression Classified
def logistic_regression_classifier(x_train, y_train):
    from sklearn.linear_model import LogisticRegression
    model = LogisticRegression()
    model.fit(x_train, y_train)
    return model

# Showing the image and the Labels(Digits Dataset)
def plot_figure():

    plt.figure()
    # zip 数据打包为元组形式
    for index, (image, label) in enumerate(zip(digits.data[0:5],digits.target[0:5])):
        plt.subplot(1, 5, index + 1)
        plt.imshow(np.reshape(image, (8, 8)))
        plt.title("Training:%i\n" % label)
    plt.show()

if __name__ == "__main__":
    digits = load_digits()

    # Print to show there are 1797 images(8*8)
    print("Image Data Shape = ", digits.data.shape)
    # Print to show There are 1797 labels (intergers form 0-9)
    print("Label Data Shape = ", digits.target.shape)

    #绘制部分图片
    plot_figure()

    # 划分训练与测试数据集
    x_train, x_test, y_train, y_test = train_test_split(digits.data, digits.target, test_size=0.25, random_state=0)

    # 搭建模型
    model = logistic_regression_classifier(x_train, y_train)

    # 预测数据
    predictions = model.predict(x_test)

    # Measuring Model Performance
    # 测试正确率，方案 ①
    accuracy = metrics.accuracy_score(y_test, predictions)
    print('accuracy: %.2f%%' % (100 * accuracy))

    # 测试正确率，方案 ②，同方案1 一致
    score = model.score(x_test, y_test)
    print("score：%.2f%%" % (100 * score))

参考文章：
https://towardsdatascience.com/logistic-regression-using-python-sklearn-numpy-mnist-handwriting-recognition-matplotlib-a6b31e2b166a