Slicing one column from a NumPy array with datatype?

Question

I have constructed a NumPy array and filled it with data:

data = np.zeros(ndata,dtype=[('datetime',np.dtype(np.datetime64)),\
    ('Value','<f8'),('Weight','<f8')])

However, now I want to do things like returning just the datetime 'column' or just the values. But this isn't a 2D NumPy array. This is a 1D array of some custom dtype, so I can't do for example:

data[:,0]

I assume this is trivial or I've fundamentally got the wrong idea about how to use NumPy arrays.

James Porter · Accepted Answer · 2013-08-05 03:32:43Z

2

A numpy datatype can be accessed just like a dictionary, so you can just do:

data[0]['datetime']

To get the datetime field of row zero or

data[:]['datetime']

To get an array of the datetime field for all rows.

answered Aug 5, 2013 at 3:32

James Porter

1,9532 gold badges19 silver badges30 bronze badges

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

2 Comments

Bogdanovist Over a year ago

Oh great, I was hoping the answer was something as trivial as this. Thanks.

James Porter Over a year ago

No problem! Often in python you can just try the easiest thing you can think of and it will work :)

Collectives™ on Stack Overflow

Slicing one column from a NumPy array with datatype?

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related