實驗01 波士頓房價預測

 

實現代碼:

 

from sklearn.linear_model import LinearRegression, SGDRegressor, Ridge, LogisticRegression
from sklearn.datasets import load_boston
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import mean_squared_error
from sklearn.externals import joblib
from sklearn.metrics import r2_score
from sklearn.neural_network import MLPRegressor

import pandas as pd
import numpy as np

lb = load_boston()
# train_test_split(train_data,train_target,test_size=0.3,random_state=5)
#train_data：待划分樣本數據
#train_target：待划分樣本數據的結果（標簽）
#test_size：測試數據占樣本數據的比例，若整數則樣本數量
#random_state：設置隨機數種子，保證每次都是同一個隨機數。若為0或不填，則每次得到數據都不一樣
#train_test_split()函數是用來隨機划分樣本數據為訓練集和測試集的，當然也可以人為的切片划分
x_train, x_test, y_train, y_test = train_test_split(lb.data, lb.target, test_size=0.2)


# 為數據增加一個維度,相當於把[1, 5, 10] 變成 [[1, 5, 10],]
y_train = y_train.reshape(-1, 1)
y_test = y_test.reshape(-1, 1)

# 進行標准化
std_x = StandardScaler()
x_train = std_x.fit_transform(x_train)
x_test = std_x.transform(x_test)

std_y = StandardScaler()
y_train = std_y.fit_transform(y_train)
y_test = std_y.transform(y_test)

# 正規方程預測
#最小二乘法線性回歸
lr = LinearRegression()
#fit_transform方法是fit和transform的結合，fit_transform(X_train) 意思是找出X_train的均值和標准差，並應用在X_train上
lr.fit(x_train, y_train)
print("r2 score of Linear regression is",r2_score(y_test,lr.predict(x_test)))

#嶺回歸
from sklearn.linear_model import RidgeCV
#嶺回歸模型
cv = RidgeCV(alphas=np.logspace(-3, 2, 100))
cv.fit (x_train , y_train)
print("r2 score of Linear regression is",r2_score(y_test,cv.predict(x_test)))

#梯度下降 用於判斷使用凸loss函數（convex loss function）的分類器
sgd = SGDRegressor()
#一個數組X（其size為[n_samples, n_features]）：保存着訓練樣本；一個數組Y：保存着訓練樣本的target值（class label）：
sgd.fit(x_train, y_train)
print("r2 score of Linear regression is",r2_score(y_test,sgd.predict(x_test)))


from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense


#基准NN
#使用標准化后的數據
seq = Sequential()
#構建神經網絡模型
#input_dim來隱含的指定輸入數據shape
seq.add(Dense(64, activation='relu',input_dim=lb.data.shape[1]))
seq.add(Dense(64, activation='relu'))
seq.add(Dense(1, activation='relu'))
seq.compile(optimizer='rmsprop', loss='mse', metrics=['mae'])
seq.fit(x_train, y_train,  epochs=300, batch_size = 16, shuffle = False)
score = seq.evaluate(x_test, y_test,batch_size=16) #loss value & metrics values
print("score:",score)
print('r2 score:',r2_score(y_test, seq.predict(x_test)))

 

 

 

運行結果:

 正規方程預測:

 

 嶺回歸結果:

 

 梯隊下降:

 

 最終結果:

 

 

 

遇到的問題及解決方法:

 

 

原因:

tensorflow 版本過高，該函數已經整合到tensorflow當中。

解決方法：

 由

from keras.models import Sequential
from keras.layers import Dense

改為:

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense