Iris 데이터 분류 - 결정 트리

Iris

Classification - Decision Tree

사용 라이브러리

import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import koreanize_matplotlib
from sklearn.datasets import load_iris
from sklearn.tree import DecisionTreeClassifier, plot_tree
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

Data Load

iris = load_iris()
iris_data = iris.data # feature만으로 된 데이터
iris_label = iris.target # label 데이터

df_iris = pd.DataFrame(data=iris_data, columns=iris.feature_names)
df_iris["label"] = iris.target
df_iris.sample(5)

	sepal length (cm)	sepal width (cm)	petal length (cm)	petal width (cm)	label
29	4.7	3.2	1.6	0.2	0
89	5.5	2.5	4.0	1.3	1
37	4.9	3.6	1.4	0.1	0
45	4.8	3.0	1.4	0.3	0
140	6.7	3.1	5.6	2.4	2

학습

X_train, X_test, y_train, y_test = train_test_split(iris_data, iris_label, test_size=0.1, random_state=42)

dt_clf = DecisionTreeClassifier(random_state=42)
dt_clf

DecisionTreeClassifier(random_state=42)

dt_clf.fit(X_train, y_train)

DecisionTreeClassifier(random_state=42)

y_predict = dt_clf.predict(X_test)

accuracy_score(y_test, y_predict)

1.0

데이터 셋의 크기가 작은것이 성능에 영향을 많이 준듯..

모델 분석

features_names = df_iris.columns.tolist()
features_names.remove("label")
features_names

['sepal length (cm)',
 'sepal width (cm)',
 'petal length (cm)',
 'petal width (cm)']

_ = sns.barplot(x=dt_clf.feature_importances_, y=features_names)

plt.figure(figsize=(12, 12))
pt = plot_tree(dt_clf, filled=True, fontsize=8, feature_names=features_names, max_depth=4)
plt.show()

)