Filter Pandas dataframe based on the list of columns and list of values

Question

I would like to be able to filter pandas using 2 lists. One list contains columns to be filtered, the second list contains which values of columns on the first list shall be selected respectively.

I have prepared an example:

import seaborn as sns
import pandas as pd

dataf = sns.load_dataset('tips')

cols = ['tip', 'sex']
vals = [1.01, 'Female']

cols2 = ['tip', 'smoker', 'day']
vals2 = [3.00, 'No', 'Sun']

# Pseudo idea of what I need
# Pseudo idea of what I need
dataf.loc[lambda d: d[cols] == vals]
# should be equal to
dataf.loc[(dataf['sex'] == 'Female') & (dataf['tip'] == 1.01)]


dataf.loc[lambda d: d[cols2] == vals2]
# should be equal to
dataf.loc[(dataf['smoker'] == 'No') & (dataf['tip'] == 3) & (dataf['day'] == 'Sun')]

I gave also an idea of what I need. However, it is very important that this should be generalizable, meaning that cols and cols2 can have a different numbers of elements within.

It will always hold that len(cols) == len(vals) ...

provide some sample data... an approach would be build a string that is valid to query() — Rob Raymond
– Rob Raymond, Commented Mar 16, 2021 at 20:31
What in case only one of the columns holds the desired value, but the other does not, e.g., [2, 'Male']? — Michael
– Michael, Commented Mar 16, 2021 at 20:42

Michael · Accepted Answer · 2021-03-16 20:52:17Z

1

This code will provide only the lines which has all the values as in the desired values array:

dataf.loc[:, cols][np.product(np.int8(dataf.loc[:, cols].isin(vals)), axis=1, dtype=bool)]

Output:


tip sex
0   1.01    Female

Cheers.

answered Mar 16, 2021 at 20:52

Michael

2,3896 gold badges27 silver badges45 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Filter Pandas dataframe based on the list of columns and list of values

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related