Thank you Anne I was doing something totally stupid at my end.
Ross On Wed, Sep 22, 2010 at 3:42 PM, Anne Archibald <[email protected]> wrote: > On 22 September 2010 16:38, Ross Williamson > <[email protected]> wrote: >> Hi everyone >> >> I suspect this is easy but I'm stuck >> >> say I have a 1D array: >> >> t = [10,11,12] >> >> and a 2D array: >> >> id = [[0,1,0] >> [0,2,0] >> [2,0,2]] >> >> In could in IDL do y = t[id] which would produce: >> >> y = [[10,11,10] >> [10,12,10] >> [12,10,12]] >> >> i.e. use the indexes in id on the lookup array t. >> >> Is there an easy way to do this in numpy? > > In [1]: t = np.array([10,11,12]) > > In [2]: id = np.array([[0,1,0], [0,2,0], [2,0,2]]) > > In [3]: t[id] > Out[3]: > array([[10, 11, 10], > [10, 12, 10], > [12, 10, 12]]) > > Anne > _______________________________________________ > NumPy-Discussion mailing list > [email protected] > http://mail.scipy.org/mailman/listinfo/numpy-discussion > -- Ross Williamson University of Chicago Department of Astronomy & Astrophysics 773-834-9785 (office) 312-504-3051 (Cell) _______________________________________________ NumPy-Discussion mailing list [email protected] http://mail.scipy.org/mailman/listinfo/numpy-discussion
