机器学习入门---逻辑斯蒂 随机梯度上升

# -*- coding: utf-8 -*-
import pandas
import numpy
#create featrue list
column_name = ['Sample code number', 'Clump Thickness', 'Uniformity of Cell Size', 'Uniformity of Cell Shape','Marginal Adhesion','Single Epithelial Cell Size','Bare Nuclei','Bland Chromatin','Normal Nucleoli','Mitoses','Class']
#获取数据
data = pandas.read_csv('https://archive.ics.uci.edu/ml/machine-learning-databases/breast-cancer-wisconsin/breast-cancer-wisconsin.data',  names=column_name)
# print(type(data))
#将数据中的空值替换为？
data = data.replace(to_replace='?', value=numpy.nan)
#删除键值存在空值的数据
data = data.dropna(how='any')

from sklearn.model_selection import train_test_split
#将数据集分为 训练集和测试集（X_train:训练数据,y_train:训练标记/特征；X_test：测试数据集，y_test:测试标记、特征）
X_train, X_test, y_train, y_test = train_test_split(data[column_name[1:10]], data[column_name[10]], test_size=0.25, random_state=33)

from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.linear_model import SGDClassifier

ss = StandardScaler()
X_train = ss.fit_transform(X_train)
X_test = ss.transform(X_test)

lr = LogisticRegression()
sc = SGDClassifier()

lr.fit(X_train, y_train)
lr_y_predict = lr.predict(X_test)

sc.fit(X_train, y_train)
sc_y_predict = sc.predict(X_test)

from sklearn.metrics import classification_report

print('逻辑斯蒂模型-准确率'.decode('utf-8'), lr.score(X_test, y_test))
print(classification_report(y_test,lr_y_predict, target_names=['良性','恶性']))

print('随机梯度上升模型-准确率'.decode('utf-8'), sc.score(X_test, y_test))
print(classification_report(y_test,sc_y_predict, target_names=['良性','恶性']))
# print(lr_y_predict)
# print(sc_y_predict)

《机器学习及实践》 阅读记录。

线性分类：

　　1.逻辑斯蒂算法；

　　2.随机梯度上升算法；

线性分类器是最常用和最基础的机器学习模型。逻辑斯蒂（LogisticRegression）和随机梯度上升（SGDClassifier）前者对参数采用精确解析的方式，计算时间场但模型性能较高，后者采用随机梯度上升估算模型参数，计算时间短但产出的模型性能略低。