pandas dataframe in a for loop with specific values

Asked Jun 09 '20 at 18:34

Active Jun 09 '20 at 18:34

Viewed 20 times

I have a sample dataframe like this:

import pandas as pd

sample_df = pd.DataFrame({'color':['red', 'green', 'red', 'red', 'yellow'],
                          'shape':['circle', 'square', 'rectangle', 'rhombus', 'line']})

And I want to get specific sample row in a new dataframe. I don't get desired output when I try to append dataframe in a for loo


value = ['red', 'green', 'yellow']
df = pd.DataFrame([])
for item in value:
    df.append(sample_df[sample_df['color'] == item].sample(1))
df.head()

output : __

How to fix it and only get one each sample of items in the value list?

asked Jun 09 '20 at 18:34

Deep

`sample_df[sample_df.color.isin(value)]` Or just for the first `sample_df[sample_df.color.isin(value)].groupby('color').first()` – yatu Jun 09 '20 at 18:37
That will return me all all the rows, since value list covers all the unique color values. But I only want one sample of each color value. – Deep Jun 09 '20 at 18:40
The second one returns one – yatu Jun 09 '20 at 18:41
Yes, second returns one sample, what if I have many such rows, and I want two samples each? – Deep Jun 09 '20 at 18:43
1

`sample_df[sample_df.color.isin(value)].groupby('color').nth([0,1])` – yatu Jun 09 '20 at 18:44
Yes, that works. And if I want it for 8 such value, all I should do is nth([0,7]) ? right? – Deep Jun 09 '20 at 18:50
This worked nth([x for x in range(8)]) – Deep Jun 09 '20 at 19:13
1

`sample_df[sample_df.color.isin(value)].groupby('color').head(8)` – yatu Jun 09 '20 at 19:20

pandas dataframe in a for loop with specific values

0 Answers0