本文中讲解的是使用sklearn实现决策树及其建模过程，包含
- 数据的清洗和数据分离
train_test_split - 采用不同的指标，基尼系数或者信息熵进行建模，使用的是
X_train和y_train- 实例化
-
fit拟合
- 预测功能：采用上面的两种实例化进行预测
y_pred = clf_gini.predict(X_test) - 结果评估
- 混淆矩阵
- 准确率
- 分类报告
`
封装成函数实现
import numpy as np
import pandas as pd
from sklearn.metrics import confusion_matrix # 混淆矩阵
from sklearn.model_selection import train_test_split # 数据分离模块
from sklearn.tree import DecisionTreeClassifier # 分类决策树
from sklearn.metrics import accuracy_score # 评价指标
from sklearn.metrics import classification_report # 生成分类结果报告模块
# 读取数据 importing data
def load_data():
balance_data = pd.read_csv('https://archive.ics.uci.edu/ml/machine-learning-'+'databases/balance-scale/balance-scale.data',sep=',',header=None) # 导入数据集，同时设置头部
print("Dataset Length", len(balance_data))
print(balance_data.head())
return balance_data
# 训练集和测试集的分离 splitting the dataset into train and test
def split_dataset(balance_data):
X = balance_data.values[:, 1:5] # 提取特征数据
y = balance_data.values[:, 0] # 提取数据标签
X_train, X_test, y_train, y_test=train_test_split(X,y,test_size=0.3,
random_state=100) # 进行数据分离
return X, y, X_train, X_test, y_train, y_test
# 使用基尼系数进行训练 training with giniIndex
def train_using_gini(X_train, y_train):
# 先建立实例，再进行fit拟合
clf_gini = DecisionTreeClassifier(criterion="gini" # 实例化
,random_state=100
,max_depth=3
,min_samples_leaf=5)
clf_gini.fit(X_train, y_train) # fit拟合
return clf_gini
# 使用信息熵进行训练 training with entropy
def train_using_entropy(X_train, y_train):
# 实例化+fit拟合
clf_entropy = DecisionTreeClassifier(criterion="entropy"
,random_state=100
,max_depth=3
,min_samples_leaf=5)
clf_entropy.fit(X_train, y_train)
return clf_entropy
# 预测功能 make predictions
def prediction(X_test, clf_object):
y_pred = clf_object.predict(X_test)
print("Predicted vlaues:")
print(y_pred)
return y_pred
# 计算准确率 calculate accuracy
def cal_accuracy(y_test, y_pred):
print("Confusion Matrix:", confusion_matrix(y_test, y_pred))
print("Accuracy:", accuracy_score(y_test, y_pred)*100)
print("Report:", classification_report(y_test, y_pred))
def main():
data = load_data()
X, y, X_train, X_test, y_train, y_test = split_dataset(data)
clf_gini = train_using_gini(X_train, y_train)
clf_entropy = train_using_entropy(X_train, y_train)
print("result using gini Index:")
y_pred_gini = prediction(X_test, clf_gini)
cal_accuracy(y_test, y_pred_gini)
print("result using Entropy:")
y_pred_entropy = prediction(X_test, clf_entropy)
cal_accuracy(y_test, y_pred_entropy)
if __name__ == "__main__":
main()
image
image
Jupyter notebook中分行实现
数据导入
# 加载UCI上的数据
data = pd.read_csv('https://archive.ics.uci.edu/ml/machine-learning-'+'databases/balance-scale/balance-scale.data',sep=',',header=None)
data.head()
image-20200115161012134
切分特征数据和数据标签
X = data.values[:, 1:5] # 切分特征数据和数据标签
y = data.values[:, 0]
数据分离
# TTS:train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.3, random_state = 100)
使用信息熵和基尼系数来建立模型
# 使用基尼系数
clf_gini = DecisionTreeClassifier(criterion = "gini",
random_state = 100,
max_depth=3,
min_samples_leaf=5)
clf_gini.fit(X_train, y_train)
# 使用信息熵
clf_entropy = DecisionTreeClassifier(criterion = "entropy",
random_state = 100,
max_depth = 3,
min_samples_leaf = 5)
clf_entropy.fit(X_train, y_train)
数据预测和评估
# 使用gini系数预测
y_pred = clf_gini.predict(X_test) # 将X_test的数据拿去进行预测，将得到的结果和y_test进行对比
confusion_matrix(y_test,y_pred) # 混淆矩阵
accuracy_score(y_test, y_pred) # 计算准确率
classification_report(y_test, y_pred) # 分类信息
image-20200115161915244
