决策树

https://www.kaggle.com/dansbecker/your-first-machine-learning-model

 1 import pandas as pd  
 2 melb_data_path = 'melb_data.csv'
 3 data1 = pd.read_csv(melb_data_path)
 4 data1.describe()
 5 data1.columns
 6 data2 = data1.dropna(axis=0)
 7 data2.describe()
 8 y = data2.Price #定义target
 9 y.describe()
10 features = ['Rooms', 'Bathroom', 'Landsize', 'Lattitude', 'Longtitude']
11 X = data2[features] #定义feature
12 X.describe()
13 X.head()
14 
15 from sklearn.tree import DecisionTreeRegressor
16 model1 = DecisionTreeRegressor(random_state=1) #选择决策树模型
17 model1.fit(X,y)  #训练模型
18 X.head()
19 model1.predict(X.head()) #使用模型对X样本前5行进行价格预测

output：

>>> X.head()

   Rooms  Bathroom  Landsize  Lattitude  Longtitude

1      2       1.0     156.0   -37.8079    144.9934

2      3       2.0     134.0   -37.8093    144.9944

4      4       1.0     120.0   -37.8072    144.9941

6      3       2.0     245.0   -37.8024    144.9993

7      2       1.0     256.0   -37.8060    144.9954

>>> model1.predict(X.head())

array([1035000., 1465000., 1600000., 1876000., 1636000.])

其他：

查看python 历史命令：    import readline; print '\n'.join([str(readline.get_history_item(i + 1)) for i in range(readline.get_current_history_length())])