Splitting Dataframe based on corresponding numpy array values

Question

I have pandas dataframe A that looks like :

    2007-12-31    50230.62
    2008-01-02    48646.84
    2008-01-03    48748.04
    2008-01-04    46992.22
    2008-01-07    46491.28
    2008-01-08    45347.72
    2008-01-09    45681.68
    2008-01-10    46430.5

Where the date column is the index. I also have an numpy array B of the same length which has element -1, 0 and 1. What is the cleanest way to split the dataframe A into 3 dataframes such that the rows with equal corresponding B elements are grouped together. For eg. if B = numpy.array([0, 0, 0, 1, 1, -1, -1, 0]) then the dataframe should be split into :

    X
    2007-12-31    50230.62
    2008-01-02    48646.84
    2008-01-03    48748.04
    2008-01-10    46430.5

    Y
    2008-01-04    46992.22
    2008-01-07    46491.28

    Z
    2008-01-08    45347.72
    2008-01-09    45681.68

Mike Müller · Accepted Answer · 2015-10-31 11:56:10Z

1

It's easy to utilize groupby from pandas, then you have the option to keep them grouped so you're not doubling your data. But you can always assign then

import numpy as np
import pandas as pd
import io

data = """    2007-12-31    50230.62
    2008-01-02    48646.84
    2008-01-03    48748.04
    2008-01-04    46992.22
    2008-01-07    46491.28
    2008-01-08    45347.72
    2008-01-09    45681.68
    2008-01-10    46430.5"""

df = pd.read_csv(io.StringIO(data), delimiter='\s+', header=None)
B = np.array([0, 0, 0, 1, 1, -1, -1, 0])

df['B'] = B

df_groups = df.groupby(['B'])

x = df_groups.get_group((0))
y = df_groups.get_group((-1))
z = df_groups.get_group((1))

The 0,-1,1 are the names based on the B value.

edited Oct 31, 2015 at 11:56

Mike Müller

86k21 gold badges174 silver badges165 bronze badges

answered Oct 31, 2015 at 4:38

Leb

16k11 gold badges58 silver badges77 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Splitting Dataframe based on corresponding numpy array values

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related