python - Pandas 为什么 pd.cut() 产生负值

Question

片段:

test = pd.DataFrame({'counts':[0,1,2,3,4,5,6,11,12,14,15]})
test['range'] = pd.cut(test.counts, [0,5,10,15], include_lowest=True)
test

输出:

    counts  range
0   0   (-0.001, 5.0]
1   1   (-0.001, 5.0]
2   2   (-0.001, 5.0]
3   3   (-0.001, 5.0]
4   4   (-0.001, 5.0]
5   5   (-0.001, 5.0]
6   6   (5.0, 10.0]
7   11  (10.0, 15.0]
8   12  (10.0, 15.0]
9   14  (10.0, 15.0]
10  15  (10.0, 15.0]

我可以得到 (0, 5.0] 而不是 (-0.001, 5.0] 吗？为什么即使我没有指定它也会出现 -0.001？

Answer 1

这是 the result of include_lowest=True internal logic .

您可以自己生成标签，方法与 pd.cuts() 在 include_lowest=False 时执行的方式相同:

In [50]: import pandas.core.algorithms as algos

In [51]: labels = pd.Categorical(pd.core.reshape.tile._format_labels(algos.unique(bins), precision=0), 
                                 ordered=True)

In [52]: labels
Out[52]:
[(0, 5], (5, 10], (10, 15]]
Categories (3, interval[int64]): [(0, 5] < (5, 10] < (10, 15]]

In [53]: test['range'] = pd.cut(test.counts, [0,5,10,15],
                                labels=labels,
                                include_lowest=True)

In [54]: test
Out[54]:
    counts     range
0        0    (0, 5]
1        1    (0, 5]
2        2    (0, 5]
3        3    (0, 5]
4        4    (0, 5]
5        5    (0, 5]
6        6   (5, 10]
7       11  (10, 15]
8       12  (10, 15]
9       14  (10, 15]
10      15  (10, 15]