决策树——过/欠拟合

本文的数据集和上一篇一样，是美国个人收入信息。

Using Decision Trees With Scikit-Learn

scikit-learn中实现了DecisionTreeClassifier分类算法以及DecisionTreeRegressor回归算法。

from sklearn.tree import DecisionTreeClassifier

# All columns have been converted to numeric.
columns = ["age", "workclass", "education_num", "marital_status", "occupation", "relationship", "race", "sex", "hours_per_week", "native_country"]

# Set random_state to 1 to keep results consistent.
clf = DecisionTreeClassifier(random_state=1)
clf.fit(income[columns], income["high_income"])

Splitting The Data Into Train And Test Sets

import numpy
import math

# Set a random seed so the shuffle is the same every time.
numpy.random.seed(1)

# Shuffle the rows.  This first permutes the index randomly using numpy.random.permutation.
# Then, it reindexes the dataframe with this.
# The net effect is to put the rows into random order.
income = income.reindex(numpy.random.permutation(income.index))

train_max_row = math.floor(income.shape[0] * .8)
train = income.iloc[:train_max_row]
test = income.iloc[train_max_row:]

Evaluating Error

from sklearn.metrics