Using df.loc, I want to see if a column is numeric, how do I do that?

Asked Oct 03 '18 at 14:33

Active Oct 03 '18 at 14:33

Viewed 62 times

I have the following DF:

Color   Name    Age   City   Value
Blue    Bob     28    Atl    0
Green   Bob     27    Chi    -
Blue    Sam     28    Atl    0

I want to change the age of all columns that have the color blue, and a numeric Value to 30. In df.loc I am not sure how to specific that I want value to be numeric. How do I do that?

asked Oct 03 '18 at 14:33

JS noob

Can you explain `I am not sure how to specific that I want value to be numeric` ? – jezrael Oct 03 '18 at 14:38
Hey, I actually just figured it out. It was pretty simple. I just did: df.Age!=np.nan – JS noob Oct 03 '18 at 14:50
Not `df.Age.notnull()` ? – jezrael Oct 03 '18 at 14:51
When I try that I get: AttributeError: 'Series' object has no attribute 'nonnull' – JS noob Oct 03 '18 at 14:53
I think `notnull()` – jezrael Oct 03 '18 at 14:53
TypeError: cannot compare a dtyped [bool] array with a scalar of type [bool] – JS noob Oct 03 '18 at 14:54
What is your full code? – jezrael Oct 03 '18 at 14:55
I am importing data from an excel sheet. Then I do: df = df.replace('-', np.nan). That might be why that is caused. – JS noob Oct 03 '18 at 14:56
And then? `df.replace('-', np.nan)` looks nice – jezrael Oct 03 '18 at 14:57
df.loc[(df.color=='Blue')&(df.Age!=np.nan),'value']='Changed' – JS noob Oct 03 '18 at 15:01
^ I do something like that afterwards. It may be the data I have in the excel sheet. – JS noob Oct 03 '18 at 15:02
1

Sorry, need Need `df.loc[(df.color=='Blue')&(df.Age.notnull()),'value']='Changed'` – jezrael Oct 03 '18 at 15:09
1

It worked. I had to restart kernel in jupyter notebook.... – JS noob Oct 03 '18 at 16:59

Using df.loc, I want to see if a column is numeric, how do I do that?

0 Answers0