📈回歸範例¶

In [ ]:

%matplotlib inline
import numpy as np
import matplotlib.pyplot as plt

%matplotlib inline import numpy as np import matplotlib.pyplot as plt

產生資料點¶

In [ ]:

np.random.seed(0)

np.random.seed(0)

In [ ]:

x = np.linspace(0, 5, 100) 

x = np.linspace(0, 5, 100)

In [ ]:

y_t = x + 1

y_t = x + 1

In [ ]:

y = y_t + 0.5 * np.random.randn(100)

y = y_t + 0.5 * np.random.randn(100)

In [ ]:

plt.scatter(x,y)
plt.plot(x,y_t,'r')

plt.scatter(x,y) plt.plot(x,y_t,'r')

將資料分成訓練集跟測試集¶

In [ ]:

from sklearn.model_selection import train_test_split

from sklearn.model_selection import train_test_split

In [ ]:

x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=0)

x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=0)

In [ ]:

len(x_train)

len(x_train)

In [ ]:

len(x_test)

len(x_test)

In [ ]:

x_train = x_train.reshape(80,1)

x_train = x_train.reshape(80,1)

In [ ]:

x_test = x_test.reshape(20,1)

x_test = x_test.reshape(20,1)

使用 sklearn 裡面的Linear Regression¶

In [ ]:

#1. 載入模型
from sklearn.linear_model import LinearRegression

#1. 載入模型 from sklearn.linear_model import LinearRegression

In [ ]:

#2. 建立模型
regr = LinearRegression()

#2. 建立模型 regr = LinearRegression()

In [ ]:

#3. 訓練模型
regr.fit(x_train, y_train)

#3. 訓練模型 regr.fit(x_train, y_train)

In [ ]:

#4. 使用模型來做預測
regr.predict([[1.3]])

#4. 使用模型來做預測 regr.predict([[1.3]])

In [ ]:

regr.predict([[2.7],[1.5]])

regr.predict([[2.7],[1.5]])

In [ ]:

regr.coef_

regr.coef_

In [ ]:

regr.intercept_

regr.intercept_

In [ ]:

y_p = regr.coef_*x + regr.intercept_

y_p = regr.coef_*x + regr.intercept_

In [ ]:

plt.plot(x,y_p,'k')
plt.plot(x,y_t,'r')

plt.plot(x,y_p,'k') plt.plot(x,y_t,'r')

In [ ]:

Y = regr.predict(x_test)

Y = regr.predict(x_test)

In [ ]:

plt.scatter(x_test,y_test)
plt.plot(x_test, Y, 'r')

plt.scatter(x_test,y_test) plt.plot(x_test, Y, 'r')