The reason of "ValueError: The truth value of ... is ambiguous" in NumPy or Pandas

The reason of 'ValueError: The truth value of ... is ambiguous' in NumPy or Pandas

I tried to do logical operation with pandas. Then it showed ValueError ...

You might see such an error when you tried to do logical operation.

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

Why does such kind of error happen ?

How can we solve it ？

So today I will introduce about "The reason of 'ValueError: The truth value of ... is ambiguous' in NumPy or Pandas".

Author

Mid-carieer engineer (AI, system). Good at Python and SQL.

Advantage to read

You can understand about the reason of 'ValueError: The truth value of ... is ambiguous' in NumPy or Pandas. Then you don't have to concern about programming with pandas.DataFrame or NumPy.ndarray.

The reason of "ValueError: The truth value of ... is ambiguous" in NumPy or Pandas

The reason of "ValueError: The truth value of ... is ambiguous" in NumPy or Pandas.

It is because you used and, or, not with NumPy or Pandas.

They are not for element-wise logical operation, but for object-wise logical operation.

Let's see sample code.

First, prepare data that contains boolean column.

DATA

import pandas as pd

data_list1 = [
    ["a",True,True],
    ["b",True,False],
    ["c",False,True],
    ["d",False,False]
]
col_list1 = ["col1","col2","col3"]
df1 = pd.DataFrame(data=data_list1, columns=col_list1)
print(df1)

#   col1   col2   col3
# 0    a   True   True
# 1    b   True  False
# 2    c  False   True
# 3    d  False  False

Then do logical operation by and, or, not.

You can see the error like below.

SAMPLE

df1["not_col2"] = not df1["col2"]
print(df1)
# ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

df1["and_col2_3"] = (df1["col2"] and df1["col3"])
print(df1)
# ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

df1["or_col2_3"] = (df1["col2"] or df1["col3"])
print(df1)
# ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

All of them occur ValueError: The truth value of ... is ambiguous.

How can we do element-wise logical operetion ?

Solution for "ValueError: The truth value of ... is ambiguous"

Object-wise operator and, or, not cause the error.

So what should we do for element-wise logical operetion ?

The solution is that using &, |, ~.

For example, use & instead of and.

Then you can get the result below.

SAMPLE

df1["and_col2_3"] = (df1["col2"] & df1["col3"])
print(df1)

#   col1   col2   col3  and_col2_3
# 0    a   True   True        True
# 1    b   True  False       False
# 2    c  False   True       False
# 3    d  False  False       False

Same as &, you can use | instead of or.

SAMPLE

df1["or_col2_3"] = (df1["col2"] | df1["col3"])
print(df1)

#   col1   col2   col3  or_col2_3
# 0    a   True   True       True
# 1    b   True  False       True
# 2    c  False   True       True
# 3    d  False  False      False

Also you can use ~ instead of not.

SAMPLE

df1["not_col2"] = ~df1["col2"]
print(df1)

#   col1   col2  not_col2
# 0    a   True     False
# 1    b   True     False
# 2    c  False      True
# 3    d   True     False

With using &, |, ~, we can do element-wise logical operetion.

Conclusion

Today I will explained about "The reason of 'ValueError: The truth value of ... is ambiguous' in NumPy or Pandas".

Points are below.

Point

and, or, not are for object-wise logical operation. So when you use them with Pandas or Numpy, they cause error.
Instead of and, or, not, you can use &, |, ~ for element-wise logical operation.