CODEWITHSUNDEEP
pythonlistpandasdataframematching
SteelyDanish
SteelyDanish

Reputation: 659

How to set values based on a list in Pandas (python)

I have a maybe a hopefully easy question that I wasn't able to find an answer on stack.

I have a a dataframe (df) and I want to set a value (some_value) if a value from the column 'month' is in a list (some_list).

e.g.

df[df['month'].isin(some_list)] = some_value

It's barfing up an AttributeError: 'int' object has no attribute 'view'.

Any helpful direction would be awesome.

[edit]:

some_list = [4,5,6,7]
some_value = 100

df.month is a value from 1 - 12

df.columns = ['datetime','weekday','hour','month','value']

I realize I also want to find out the index of the rows which isin some_list then use those indices to set the value of another column ('value') to some_value. Apologies for not writing that originally.

Upvotes: 3

Views: 18863

Answers (1)

Marius
Marius

Reputation: 60070

Your question still doesn't seem to have enough information to find the real problem. This quick example shows that your attempt can work just fine:

import pandas as pd
df = pd.DataFrame({'x': [4, 5, 6], 'month': [1, 2, 3]})
some_list = [2, 3]
df[df['month'].isin(some_list)] = 99
df
Out[13]: 
   month   x
0      1   4
1     99  99
2     99  99

...suggesting that your problem is more likely because you've mixed up the types of your variables. Currently the only thing I can suggest is only doing the assignment to specific columns, as you may be trying to assign an int value to a datetime column or something, e.g.:

df = pd.DataFrame({'x': [4, 5, 6], 'month': [1, 2, 3]})
some_list = [2, 3]
df.loc[df['month'].isin(some_list), 'x'] = 99
df
Out[14]: 
   month   x
0      1   4
1      2  99
2      3  99

Upvotes: 15

Related Questions