Hi All,
Just noting an oddity when the diagonal method is called on arrays with ndim
2
In [1]: a = arange(8).reshape(2,2,2)
In [2]: a.diagonal()
Out[2]:
array([[0, 6],
[1, 7]])
The diagonal is taken over the first two indices while the rightmost indices
vary the slowest. I would expect things to be the other way around. In
particular, it would work better with the arrays of matrices proposed by
Tim. I suppose another logical extrapolation from two dimensions would be to
take the "diagonal" on all three indices. But the current behaviour strikes
me as a bit unexpected.
Chuck
_______________________________________________
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
