29 Mar
2007
29 Mar
'07
5:24 p.m.
On Tue, 27 Mar 2007, Zachary Pincus apparently wrote:
Now, Bill offers up a different suggestion: indexing M yields neither a matrix nor an array, but a class that operates more or less like an array, except insofar as it interacts with other matrix objects, or other objects of similar classes. I'm interested in hearing more about this, what trade-offs or other compromises it might involve.
So would you agree that in this case M[0] and M[0,:] would have to be different things? Or would you then want even such numpy indexing not to produce matrices?? Cheers, Alan Isaac