Python决策树适合方法给出浮动错误

时间:2017-05-17 10:07:11

标签: python decision-tree

我正在使用决策树分类器。 我的X(特征变量)看起来像一个字典:

{'location05': ['4425', '4436', '4425'],
 'location04': ['425', '4436', '4425','232'],
 'location03': ['4425', '4436']  }

我的Y(目标变量)只是二进制 现在,当我把它们放入适合的方法时,我收到了错误:

  

float()参数必须是字符串或数字

有任何帮助吗?

我的代码如下:

from sklearn.tree import DecisionTreeClassifier
import numpy as np
import csv
from sklearn import tree

reader = csv.reader(open("data.csv", "rb"))
next(reader, None)
location_dict = dict()
target = []
feat = []
for line in reader:
    feat.append(line[2])
    target.append(line[3])

    if line[0] in location_dict:
        location_dict[line[0]].append(line[2])
    else:
        location_dict[line[0]] = [line[2]]
X = location_dict
y = target
clf = tree.DecisionTreeClassifier()
clf = clf.fit(X, y)

0 个答案:

没有答案