how to convert pandas Series of Numpy object into a Numpy matrix?

Question

I find it hard to convert pandas Series of size M each containing a numpy arrays each of size N into a matrix/numpy array/dataframe of size MxN

Example:

import pandas as pd
import numpy as np
from scipy import stats
d = pd.DataFrame({'grp': np.random.randint(1, 10, 1000), 'x':np.random.rand(1000,)})
s = d.groupby('grp')['x'].apply(lambda x: stats.gaussian_kde(x.values, bw_method = .01).evaluate(np.linspace(0,1,100)))

The output I get is of type Series where the type of the entries are numpy.ndarray. How do I convert this to be size of 10 (groups) times 100 (evaluation bins)?

mrcl · Accepted Answer · 2014-10-23 22:46:53Z

2

You can use np.vstack for that

sArray = np.vstack(s)

But in your case your Series is (9,100). If you want to make it (10,100) you need to change the line.

d = pd.DataFrame({'grp': np.random.randint(1, 10, 1000), 'x':np.random.rand(1000,)})

to

d = pd.DataFrame({'grp': np.random.randint(0, 10, 1000), 'x':np.random.rand(1000,)})

or

d = pd.DataFrame({'grp': np.random.randint(1, 11, 1000), 'x':np.random.rand(1000,)})

Not sure what you want to do there.

Cheers

answered Oct 23, 2014 at 22:46

mrcl

2,19014 silver badges29 bronze badges

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

Collectives™ on Stack Overflow

how to convert pandas Series of Numpy object into a Numpy matrix?

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