On Di, 2015-11-17 at 13:49 -0500, Neal Becker wrote: > Robert Kern wrote: > > > On Tue, Nov 17, 2015 at 3:48 PM, Neal Becker <ndbeck...@gmail.com> wrote: > >> > >> I have an array of shape > >> (7, 24, 2, 1024) > >> > >> I'd like an array of > >> (7, 24, 2048) > >> > >> such that the elements on the last dimension are interleaving the > >> elements from the 3rd dimension > >> > >> [0,0,0,0] -> [0,0,0] > >> [0,0,1,0] -> [0,0,1] > >> [0,0,0,1] -> [0,0,2] > >> [0,0,1,1] -> [0,0,3] > >> ... > >> > >> What might be the simplest way to do this? > > > > np.transpose(A, (-2, -1)).reshape(A.shape[:-2] + (-1,)) > > I get an error on that 1st transpose: >
Transpose needs a slightly different input. If you look at the help, it should be clear. The help might also point to np.swapaxes, which may be a bit more straight forward for this exact case. > here, 'A' is 'fftouts' > > print (fftouts.shape) > print (np.transpose (fftouts, (-2,-1)).shape) > > (4, 24, 2, 1024) <<< fftouts.shape prints this > Traceback (most recent call last): > File "test_uw2.py", line 194, in <module> > run_line (sys.argv) > File "test_uw2.py", line 190, in run_line > run (opt) > File "test_uw2.py", line 103, in run > print (np.transpose (fftouts, (-2,-1)).shape) > File "/home/nbecker/.local/lib/python2.7/site- > packages/numpy/core/fromnumeric.py", line 551, in transpose > return transpose(axes) > ValueError: axes don't match array > > > > >> ------------ > >> A different question, suppose I just want to stack them > >> > >> [0,0,0,0] -> [0,0,0] > >> [0,0,0,1] -> [0,0,1] > >> [0,0,0,2] -> [0,0,2] > >> ... > >> [0,0,1,0] -> [0,0,1024] > >> [0,0,1,1] -> [0,0,1025] > >> [0,0,1,2] -> [0,0,1026] > >> ... > > > > A.reshape(A.shape[:-2] + (-1,)) > > > > -- > > Robert Kern > > > _______________________________________________ > NumPy-Discussion mailing list > NumPy-Discussion@scipy.org > https://mail.scipy.org/mailman/listinfo/numpy-discussion >
signature.asc
Description: This is a digitally signed message part
_______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org https://mail.scipy.org/mailman/listinfo/numpy-discussion
