0
import pandas as pd; import numpy as np;
df = {'id': [1, 2, 3, 4, 5, 6],
   'created_at': ['2020-02-01', '2020-02-02', '2020-02-02', '2020-02-02', '2020-02-03','2020-02-02'],
   'type': ['red', np.nan, np.nan, 'blue', 'yellow', np.nan]}

df = pd.DataFrame (df, columns = ['id', 'created_at','type'])

If created_at=2020-02-01 and id=2 or 6 drop row else don't drop. I want to obtain this output;

id created_at type
1 2020-02-01 red
3 2020-02-01 NaN
4 2020-02-01 blue
5 2020-02-03 yellow

i.e. I don't want to drop all rows with nan value.

Ynjxsjmh
  • 28,441
  • 6
  • 34
  • 52
bengisu yilmaz
  • 21
  • 5

1 Answers1

1

You would need to use boolean indexing:

# is the date 2020-02-01?
m1 = df['created_at'].eq('2020-02-01')
# is the id 2 or 6?
m2 = df['id'].isin([2, 6])

# keep if NOT both conditions are matched
out = df[~(m1&m2)]

Alternatively:

# is the date NOT 2020-02-01?
m1 = df['created_at'].ne('2020-02-01')
# is the id NOT 2 or 6?
m2 = ~df['id'].isin([2, 6])

# keep if either condition is matched
out = df[m1|m2]
mozway
  • 194,879
  • 13
  • 39
  • 75