2.1.14.2.Decision trees and Random Forests with Python

1. 匯入基本的library

• pandas, numpy, matplotlib, seaborn

    import pandas as pd
    import numpy as np
    import matplotlib.pyplot as plt
    import seaborn as sns

• 將圖表直接嵌入到Notebook之中

    %matplotlib inline

2.讀取資料並了解資料

• 讀取資料

  • 由df.head可知變因有Age, Number, Start, 結果為Kyphosis

     df = pd.read_csv('kyphosis.csv')
     df.head()

    

• 先了解資料欄位的型別以及變數的型態, 由pd.info()可以知道這份資料有4個欄位: 有3筆屬於int64, 1筆屬於object

   df.info()

  

• 視覺化資料以了解每個因子間的相關性

  • seaborn pairpot

    sns.pairplot(data = df, hue='Kyphosis')

    

3.使用Skikit-learn library

• 首先介紹train_test_split, 這個函式可以隨機劃分訓練集和測試集

    from sklearn.cross_validation import train_test_split
    X = df_feat
    Y = df["TARGET CLASS"]
    X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.4, random_state=101)

4.使用Decision tree classifier

• import Decision tree classifier

    from sklearn.tree import DecisionTreeClassifier
    dtree = DecisionTreeClassifier()

• 訓練模型

   dtree.fit(X_train, y_train)

• 預測

    pred = dtree.predict(X_test)

• 評估模型的精度

  • confusion_matrix, classification_report

    from sklearn.metrics import classification_report, confusion_matrix
    print(classification_report(y_test, pred))
    print(confusion_matrix(y_test, pred))

    

5.使用Random tree forest

• import Random tree forest

    from sklearn.ensemble import RandomForestClassifier

• 訓練模型

    rfc = RandomForestClassifier(n_estimators = 200)
    rfc.fit(X_train, y_train)

• 預測

    rfc_pred = rfc.predict(X_test)

• 評估模型的精度

    rfc_pred = rfc.predict(X_test)
    print(classification_report(y_test, rfc_pred))
    print(confusion_matrix(y_test, rfc_pred))

  

6.視覺化決策樹