熊猫"变换(设置)"引发异常

Question

我正在尝试按set进行转换，但我得到了一个例外。转换适用于"sum"和许多其他聚合函数，但不适用于set或list。

>>> import pandas as pd
>>> df = pd.DataFrame({"a":[1,2,1,], "b":[1,1,2]})
>>> df
a  b
0  1  1
1  2  1
2  1  2
>>> df.groupby("a").b.transform(set)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/Users/avloss/conda/lib/python3.7/site-packages/pandas/core/groupby/generic.py", line 1032, in transform
s = klass(res, indexer)
File "/Users/avloss/conda/lib/python3.7/site-packages/pandas/core/series.py", line 282, in __init__
"{0!r} type is unordered" "".format(data.__class__.__name__)
TypeError: 'set' type is unordered

我期待的是：

a      b   
0  1  {1, 2}
1  2     {1}
2  1  {1, 2}

Answer 1

我能想到的最简单的方法是aggregate为set并将其映射回来

df['new_col'] = df['a'].map(df.groupby('a')['b'].agg(set))
print(df)

---

a  b new_col
0  1  1  {1, 2}
1  2  1     {1}
2  1  2  {1, 2}

Answer 2

我们有unique，请注意unique甚至会保留您输入的原始顺序

df.groupby('a').b.transform('unique')
0    [1, 2]
1       [1]
2    [1, 2]
Name: b, dtype: object

如果上述不起作用，则更新 ，请在下面尝试

df.groupby('a').b.transform(lambda x : [set(x)]*len(x))
0    {1, 2}
1       {1}
2    {1, 2}
Name: b, dtype: object