第7章 buildtree类型错误

[xjr7670]

请问一下，中文版P151，在执行

tree=treepredict.buildtree(treepredict.my_data)

这里，用python3不会报类型错误吗？
我直接按书中的代码执行是会报错的，用你这里的代码也是。一样的错误：

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-27-d4c8cc29c32a> in <module>()
----> 1 treepredict.buildtree(treepredict.my_data)

/home/cavin/Code/BookPractice/Python/ProgrammingCollectiveIntelligence/chapter7/treepredict.py in buildtree(rows, scoref)
    105         for row in rows:
    106             column_values[row[col]] = 1
--> 107         # 接下来根据这一列中的每个值，尝试对数据集进行拆分
    108         for value in column_values.keys():
    109             (set1, set2) = divideset(rows, col, value)

/home/cavin/Code/BookPractice/Python/ProgrammingCollectiveIntelligence/chapter7/treepredict.py in divideset(rows, column, value)
     39         print(row[column], value)
     40         split_function = lambda row: row[column] == value
---> 41 
     42     # 将数据集拆分成两个集合，并返回
     43     set1 = [row for row in rows if split_function(row)]

/home/cavin/Code/BookPractice/Python/ProgrammingCollectiveIntelligence/chapter7/treepredict.py in <listcomp>(.0)
     39         print(row[column], value)
     40         split_function = lambda row: row[column] == value
---> 41 
     42     # 将数据集拆分成两个集合，并返回
     43     set1 = [row for row in rows if split_function(row)]

/home/cavin/Code/BookPractice/Python/ProgrammingCollectiveIntelligence/chapter7/treepredict.py in <lambda>(row)
     34     split_function = None
     35     if isinstance(value, int) or isinstance(value, float):
---> 36         print(row[column], value)
     37         split_function = lambda row: row[column] >= value
     38     else:

TypeError: '>=' not supported between instances of 'str' and 'int'