Pandas DataFrame | ne method
Start your free 7-days trial now!
Pandas DataFrame.ne(~) method returns a DataFrame of booleans where True indicates an entry that is not equal (!=) to the specified value.
Missing values (NaN) are considered to be distinct, that is, NaN!=NaN will evaluate to True.
Parameters
1. otherlink | scalar or sequence or Series or DataFrame
The value(s) to compare with.
2. axislink | int or string | optional
Whether to compare each row or column:
Axis | Description |
|---|---|
| Compare each column. |
| Compare each row. |
By default, axis="columns".
3. level | int or string | optional
The levels to perform comparison on. This is only relevant if your source DataFrame is a multi-index.
Return Value
A DataFrame of booleans.
Examples
Consider the following DataFrame:
df = pd.DataFrame({"A":[3,4],"B":[5,6]})df
A B0 3 51 4 6
Passing in a acalar
To check for values not equal to 5 in the DataFrame:
df.ne(5)
A B0 True False1 True True
Comparing rows
By default, axis=1, which means that passing in a sequence will result in a comparison with each row:
df.ne([4,5]) # axis=1
A B0 True False1 False True
Here, we are comparing each row of the source DataFrame with [4,5]. This means that we are performing the following pair-wise comparisons:
(row one) [3,5] != [4,5] = [True, False](row two) [4,6] != [4,5] = [False, True]
We show df again for your reference:
df
A B0 3 51 4 6
Comparing columns
By setting axis=0, we can compare each column with the specified sequence:
df.ne([4,5], axis=0)
A B0 True True1 True True
Here, we're performing the following pair-wise comparisons:
(column A) [3,4] != [4,5] = [True, True](column B) [5,6] != [4,5] = [True, True]
