鸢尾花数据集建立决策树（调用sklearn库中的决策树模型）

from sklearn.datasets import load_iris
from sklearn import tree
import pandas as pd
import pydotplus
import numpy as np
from sklearn.metrics import accuracy_score,recall_score
from sklearn.model_selection import train_test_split,cross_val_score
from sklearn.model_selection import GridSearchCV
from sklearn.metrics import classification_report
list1=['Sepal.Length', 'Sepal.Width', 'Petal.Length', 'Petal.Width']
list2=['setosa','versicolor','virginica']
iris = pd.read_csv('iris.csv', usecols=[1, 2, 3, 4, 5])
list=[]
for i in range(0,150):
    if(iris['Species'][i]=="setosa"):
        list.append(0)
    elif(iris['Species'][i]=="versicolor"):
        list.append(1)
    else:
        list.append(2)
list=np.array(list)
iris.insert(5,"classification",list)
iris.info()
iris_data= iris[['Sepal.Length', 'Sepal.Width', 'Petal.Length', 'Petal.Width']]
iris_target=iris[['classification']]
iris_data=np.array(iris_data)
iris_target=np.array(iris_target)
X_train,X_test, y_train, y_test =train_test_split(iris_data,iris_target,test_size=0.3, random_state=0)
clf= tree.DecisionTreeClassifier(criterion="entropy",max_depth=4,max_leaf_nodes=3,min_samples_leaf=2,min_samples_split=2)
clf.fit(X_train,y_train)
print(clf.get_params())
pred5=clf.predict(X_train)
print(accuracy_score(y_train,pred5))

pred1=clf.predict(X_test)
print(accuracy_score(y_test,pred1))


clf1=tree.DecisionTreeClassifier()
gridparams=[
    {
        "max_depth": [1,2,3,4,5,6,7],
        "max_leaf_nodes": [3,4,5,6,7,8,9],
        "min_samples_leaf": [1,2,3,4,5,6,7,8,9],
        "min_samples_split": [2,3,4,5,6,7,8,9]
    }
]
grid_search = GridSearchCV(clf1, gridparams,cv=3,scoring="accuracy")
grid_search.fit(X_train,y_train)


# pred2=grid_search.predict(X_test)
# print(accuracy_score(y_test,pred2))
best_params = grid_search.best_params_
best_scores = grid_search.best_score_
best_estimator = grid_search.best_estimator_
print(best_params)
print(best_scores)
print(best_estimator)



clf2= tree.DecisionTreeClassifier(criterion="gini",max_depth=3,max_leaf_nodes=5,min_samples_leaf=1,min_samples_split=2)
clf2.fit(X_train,y_train)
pred2=clf2.predict(X_test)
print(accuracy_score(y_test,pred2))
#print(recall_score(y_test,pred2))

dot_data = tree.export_graphviz(clf, out_file=None, 
                         feature_names=list1,  
                         class_names=list2,  
                         filled=True, rounded=True,  
                         special_characters=True)
graph = pydotplus.graph_from_dot_data(dot_data)
from IPython.display import Image
Image(graph.create_png())

dot_data1 = tree.export_graphviz(clf2, out_file=None, 
                         feature_names=list1,  
                         class_names=list2,  
                         filled=True, rounded=True,  
                         special_characters=True)
graph1 = pydotplus.graph_from_dot_data(dot_data1)
from IPython.display import Image
Image(graph1.create_png())