Use numpy to read a two-column csv file

Question

I was curious whether there is a nicer way to do this. I have a csv file with two columns (which is fairly common), for example the 1st column is the time stamp and the second column is the data.

# temp.csv
0., 0.
1., 5.
2., 10.
3., 15.
4., 10.
5., 0.

And then I want to read this temp.csv file:

import numpy as np

my_csv = np.genfromtxt('./temp.csv', delimiter=',')
time = my_csv[:, 0]
data = my_csv[:, 1]

This is entirely fine, but I am just curious whether there is a more elegant way to do this, since this is a fairly common practice.

Thank you!

-Shawn

Answer 1

You can do:

my_csv = np.genfromtxt('./temp.csv', delimiter=',')
time, data = my_csv.transpose()

or the one liner:

time, data = np.genfromtxt('./temp.csv', delimiter=',').transpose()

or another one liner where genfromtxt does the transposition:

time, data = np.genfromtxt('./temp.csv', delimiter=',', unpack=True)

Are they more elegant? That's up to the reader.

Answer 2

You can use the unpack argument to have genfromtxt return the transpose of the array. Then you can do your assignments like this:

In [3]: time, data = genfromtxt('temp.csv', unpack=True, delimiter=',')

In [4]: time
Out[4]: array([ 0.,  1.,  2.,  3.,  4.,  5.])

In [5]: data
Out[5]: array([  0.,   5.,  10.,  15.,  10.,   0.])

Answer 3

Use pandas dataframes if you want nice ways for working with csv type tables.'

http://pandas.pydata.org/pandas-docs/dev/dsintro.html