sklearn Read custom datasets

import csv
from sklearn.utils import Bunch

#  Read watermelon dataset 
def readWatermelonDataSet():
    FeatureNames = []
    FeatureList = []
    LabelList = []

    ifile = open("E:\My Word\study\RL0314\data.csv", "r")
    reader = csv.reader(ifile)
    cnt = 0
    for row in reader:
        if cnt == 0:  #  Read attribute name 
            headers = row
            FeatureNames = headers[1:len(headers) - 1]
            # print(FeatureNames)
        else:  #  Read data and labels 
            headers = row
            FeatureList.append(headers[1:len(headers) - 1])
            LabelList.append(headers[len(headers) - 1])
        cnt = cnt + 1
    print(FeatureNames)
    print(FeatureList)
    print(LabelList)

    return Bunch(
        data=FeatureList,
        target=LabelList,
        feature_names=FeatureNames,
    )

Be careful ： If you want to use it directly sklearn Subsequent algorithm , The data set should be numeric data , If you add other columns of watermelon dataset, you will report errors later , Data preprocessing is needed .

The data set used here is like this ：

Complete decision tree generation code ：

import csv
from sklearn.utils import Bunch
from sklearn import tree
from sklearn.model_selection import train_test_split
import pandas as pd
import graphviz
import os


#  Read watermelon dataset 
def readWatermelonDataSet():
    FeatureNames = []
    FeatureList = []
    LabelList = []

    ifile = open("E:\My Word\study\RL0314\data.csv", "r")
    reader = csv.reader(ifile)
    cnt = 0
    for row in reader:
        if cnt == 0:  #  Read attribute name 
            headers = row
            FeatureNames = headers[1:len(headers) - 1]
            # print(FeatureNames)
        else:  #  Read data and labels 
            headers = row
            FeatureList.append(headers[1:len(headers) - 1])
            LabelList.append(headers[len(headers) - 1])
        cnt = cnt + 1
    print(FeatureNames)
    print(FeatureList)
    print(LabelList)

    return Bunch(
        data=FeatureList,
        target=LabelList,
        feature_names=FeatureNames,
    )


def main():
    watermelon = readWatermelonDataSet()  #  Watermelon data 
    pd.concat([pd.DataFrame(watermelon.data), pd.DataFrame(watermelon.target)], axis=1)
    Xtrain, Xtest, Ytarin, Ytest = train_test_split(watermelon.data, watermelon.target, test_size=0.3)  #  Test set 30% Training set 70%

    """ Build a model """
    clf = tree.DecisionTreeClassifier(criterion="entropy")  #  Instantiation , Classification tree 
    clf = clf.fit(Xtrain, Ytarin)
    score = clf.score(Xtest, Ytest)
    score

    dot_data = tree.export_graphviz(clf
                                    , feature_names=watermelon.feature_names
                                    , class_names=[" Good melon ", " Bad melon "]
                                    , filled=True
                                    , rounded=True
                                    , special_characters=True
                                    , fontname="Microsoft YaHei")

    graph = graphviz.Source(dot_data)
    os.environ["PATH"] += os.pathsep + 'D:/DiyProgram/graphviz/bin/'
    graph.render("watermelon1", view=True)


if __name__ == "__main__":
    main()

Running results ：