Filtering multiple columns with conditions in Pandas-CodePudding

I am trying to filter multiple column values by conditions but I am wondering if there's a faster or concise way to do it without putting down one column by one column, like grouping the conditions or map().

dt = pd.DataFrame(
    {
        "Num1": [1, -2, 3, -4, 5],
        "Num2": [-1, 2, -3, 4, 5],
        "Num3": [1, -2, 3, 4, -5],
        "Zum1": [1, 2, 3, -4, 5],
        "Num5": [1, -2, 3, -4, 5]
    }
); dt

# current solution:
dt[(abs(dt["Num1"] > 2) | abs(dt["Num2"] > 2) | abs(dt["Num3"] > 2)) & (dt['Zum1'] > 2)]

# expected outcome
   Num1  Num2  Num3  Zum1  Num5
2     3    -3     3     3     3
4     5     5    -5     5     5

CodePudding user response：

If there is multiple conditions for compare by same value is possible use:

dt = dt[dt[['Num1','Num2','Num3']].abs().gt(2).any(1) & (dt['Zum1'] > 2)]
print (dt)
   Num1  Num2  Num3  Zum1  Num5
2     3    -3     3     3     3
4     5     5    -5     5     5

Or if need filter all column with Num use DataFrame.filter:

dt = dt[dt.filter(like='Num').abs().gt(2).any(1) & (dt['Zum1'] > 2)]