如何结合多索引和数据透视在熊猫中创建真值表？

Question

I have the following Pandas DataFrame, where column Z can have an unknown number of possible values eg A,B,C,D,E ... :

X Y Z
-----
1 1 A
1 1 B
1 1 C
1 2 A
2 1 B
2 1 C
2 2 B

I would like to pivot on column Z and have multi-index on columns X,Y, to create a truth table like so:

X Y A B C
---------
1 1 T T T
1 2 T F F
2 1 F T T
2 2 F T F

How do I do this in pandas ?

Answer 1

You can also create a value column of Trues and then use pivot_table() :

df["Value"] = True   
df.pivot_table("Value", ["X", "Y"], "Z", fill_value=False).reset_index()

Answer 2

The following code does it (almost):

pd.get_dummies(df,'Z').groupby(['X','Y']).max()

The following is closer to the exact output you asked for:

(pd.get_dummies(df,'Z')
 .groupby(['X','Y'])
 .max()
 .astype(bool)
 .applymap(lambda s: str(s)[0])
 .rename(columns=lambda v: v.replace('Z_','')))