100 Days Of ML Code：Day2-Simple Linear Regression（一元线性回归）

100天机器学习挑战汇总文章链接在这儿。

目录

Step 1：数据预处理

Step 2：对训练集应用简单的线性回归模型

Step 3：预测结果

Step 4：绘图

最后：全部代码

---

Step 1：数据预处理

这一段参见Day1的内容，数据预处理。

import pandas as pd
df = pd.read_csv('studentscores.csv')
# print(df)
X = df.iloc[:, :1].values
# X = df.iloc[:, 0].values 不能是这句，因为这样得到的是1D的向量，而后面的regressor.fit函数必须是2D的输入
Y = df.iloc[:, 1].values
print(X)
# print(Y)

# from sklearn.preprocessing import Imputer
# imp = Imputer(missing_values='NaN', strategy='mean', axis=0, verbose=0, copy=True)
# imp.fit(X)
# X[:, 1:] = imp.transform(X)
# print(X)

from sklearn.cross_validation import train_test_split
X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.20, random_state=0)
# print(X_train)

注意此处特征虽然只有学习时间这一个（一元线性回归），但是X必须是2D的（如下），否则后面的regressor.fit函数会出错。

[[ 2.5]
 [ 5.1]
 [ 3.2]
 [ 8.5]
 [ 3.5]
 [ 1.5]
 [ 9.2]
 [ 5.5]
 [ 8.3]
 [ 2.7]
 [ 7.7]
 [ 5.9]
 [ 4.5]
 [ 3.3]
 [ 1.1]
 [ 8.9]
 [ 2.5]
 [ 1.9]
 [ 6.1]
 [ 7.4]
 [ 2.7]
 [ 4.8]
 [ 3.8]
 [ 6.9]
 [ 7.8]]

Step 2：对训练集应用简单的线性回归模型

LinearRegression的指导页面在这儿，还可参考这篇文章，介绍的很详细。

from sklearn.linear_model import LinearRegression
regressor = LinearRegression()
regressor = regressor.fit(X_train, Y_train)
print(regressor.coef_)
print(regressor.intercept_)

其中coef_存放回归系数，intercept_则存放截距。他们结果分别是：

[ 9.91065648]
2.01816004143

Step 3：预测结果

Y_pred = regressor.predict(X_test)
print(Y_pred)

打印的结果是：

[ 16.88414476  33.73226078  75.357018    26.79480124  60.49103328]

Step 4：绘图

import matplotlib.pyplot as plt
plt.scatter(X_train, Y_train, color = 'red')
plt.plot(X_train, regressor.predict(X_train), color ='blue')
plt.scatter(X_test, Y_test, color = 'red')
plt.plot(X_test, regressor.predict(X_test), color ='blue')
plt.show()

最后：全部代码

最后贴一下整个过程中全部的代码：

import pandas as pd
df = pd.read_csv('studentscores.csv')
# print(df)
X = df.iloc[:, :1].values
# X = df.iloc[:, 0].values 不能是这句，因为这样得到的是1D的向量，而后面的regressor.fit函数必须是2D的输入
Y = df.iloc[:, 1].values
# print(X)
# print(Y)

# from sklearn.preprocessing import Imputer
# imp = Imputer(missing_values='NaN', strategy='mean', axis=0, verbose=0, copy=True)
# imp.fit(X)
# X[:, 1:] = imp.transform(X)
# print(X)

from sklearn.cross_validation import train_test_split
X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.20, random_state=0)
# print(X_train)

from sklearn.linear_model import LinearRegression
regressor = LinearRegression()
regressor = regressor.fit(X_train, Y_train)
# print(regressor.coef_)
# print(regressor.intercept_)

Y_pred = regressor.predict(X_test)
# print(Y_pred)

import matplotlib.pyplot as plt
plt.scatter(X_train, Y_train, color = 'red')
plt.plot(X_train, regressor.predict(X_train), color ='blue')
plt.scatter(X_test, Y_test, color = 'red')
plt.plot(X_test, regressor.predict(X_test), color ='blue')
plt.show()