To whom it may concern: 2 issues:
1. In MATLAB, we can form an nxn FFT matrix by doing fft(eye(n)). In Julia, doing fft(eye(n)) doesn't seem to be giving me the same result. 2. I am actually interested in randomly sampling the rows of a FFT matrix, and doing matrix-vector multiples with only those rows. I was wondering if there was a way to use plan_fft to get the nlogn flop speed using plan_fft in this case. Thanks! Ed
