I have 19 columns (q1 to q19) in a DataFrame 'users' I want to convert from float to int. Instead of manually typing it out is there a way to automate the process?
Code I have so far is:
users.q1 = users.q1.astype(int)
-
Sorry I forgot to mention there are other columns in the dataframe I want to keep as other datatypesEduardo Lucero– Eduardo Lucero2018-06-12 04:08:21 +00:00Commented Jun 12, 2018 at 4:08
Add a comment
|
3 Answers
Pass a list of the columns that you want to alter:
import pandas as pd
df1 = pd.DataFrame({'A': [1, 2, 3], 'B': [2, 3, 8], 'C':[5, 9, 12]})
df1[['A', 'B']] = df1[['A', 'B']].astype('float')
print(df1)
Output:
A B C
0 1.0 2.0 5
1 2.0 3.0 9
2 3.0 8.0 12
Instead of altering line for line, we can simplify to one line for all of the columns that need to be changed.
4 Comments
U13-Forward
One thing,
float in the astype does not need to be a string
Simeon Ikudabo
@U8-Forward I am fairly certain that in numpy it does have to be a string, which is why I do so in Pandas. That may not be accurate, but I vaguely remember it having to be a string a few years ago, and I assumed it was the same in pandas. I could be wrong.
U13-Forward
Ok but at least i know now that you can just do
.astype(float)Simeon Ikudabo
@U8-Forward Thanks. I will test if you can do that in numpy. I recall getting a syntax error in Numpy when I attempted to pass float in the "dtype" argument a few years ago. So I have figured it was the same in Pandas ever since. Thanks for the knowledge. I just thought that you had to pass a string.
you can try this
df.loc[:, 'q1':'q19'] = df.loc[:, 'q1':'q19'].astype(int)
Comments
Try this:
for i in users.columns:
if <condition>:
users[i] = users[i].astype(int)
Comments
