《Web安全之机器学习入门》笔记：第八章 8.4 逻辑回归算法识别mnist验证码

        本小节是通过使用逻辑回归算法对mnist数据集的数字识别，效果只能说勉强凑合，不过比7.8节的nb算法好一些。

        1.源码修改

        作者的代码会报错以及报警

        (1)报错

Traceback (most recent call last):
  File "C:/Users/liujiannan/PycharmProjects/pythonProject/Web安全之机器学习入门/code/8-3.py", line 15, in <module>
    training_data, valid_data, test_data=load_data()
  File "C:/Users/liujiannan/PycharmProjects/pythonProject/Web安全之机器学习入门/code/8-3.py", line 10, in load_data
    training_data, valid_data, test_data = pickle.load(fp)
UnicodeDecodeError: 'ascii' codec can't decode byte 0x90 in position 614: ordinal not in range(128)

        修改方法

def load_data():
    with gzip.open('../data/MNIST/mnist.pkl.gz') as fp:
        training_data, valid_data, test_data = pickle.load(fp, encoding="bytes")
    return training_data, valid_data, test_data

        （2）报警

C:\ProgramData\Anaconda3\lib\site-packages\sklearn\linear_model\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.
  FutureWarning)
C:\ProgramData\Anaconda3\lib\site-packages\sklearn\linear_model\logistic.py:469: FutureWarning: Default multi_class will be changed to 'auto' in 0.22. Specify the multi_class option to silence this warning.
  "this warning.", FutureWarning)

源码修改

    logreg = linear_model.LogisticRegression(C=1e5, solver='liblinear', multi_class='ovr')

        2.完整代码

        基于原作者代码修改可运行在python3环境的源码：

# -*- coding:utf-8 -*-

from sklearn import model_selection
from sklearn.naive_bayes import GaussianNB
import pickle
import gzip


def load_data():
    with gzip.open('../data/MNIST/mnist.pkl.gz') as fp:
        training_data, valid_data, test_data = pickle.load(fp, encoding="bytes")
    return training_data, valid_data, test_data


if __name__ == '__main__':
    training_data, valid_data, test_data=load_data()
    x1,y1=training_data
    x2,y2=test_data
    clf = GaussianNB()
    clf.fit(x1, y1)
    score = model_selection.cross_val_score(clf, x2, y2, scoring="accuracy")
    print(score)
    print(score.mean())

        不过，我们看一下，交叉验证使用的是x2和y2，相当于用x2和y2既训练又测试得到的结果。

        3.运行结果

[0.76482924 0.8529853  0.8639231 ]
0.8272458792084002

本例虽然作者示例有点问题，但是实际上逻辑回归测试mnist图片效果有很多，很容易做到90%以上，这里只是展示一种用法，不要太较真。