pyfftw

问题 I have a nD array, say of dimensions: (144, 522720) and I need to compute its FFT. PyFFTW seems slower than numpy and scipy , that it is NOT expected. Am I doing something obviously wrong? Below is my code import numpy import scipy import pyfftw import time n1 = 144 n2 = 522720 loops = 2 pyfftw.config.NUM_THREADS = 4 pyfftw.config.PLANNER_EFFORT = 'FFTW_ESTIMATE' # pyfftw.config.PLANNER_EFFORT = 'FFTW_MEASURE' Q_1 = pyfftw.empty_aligned([n1, n2], dtype='float64') Q_2 = pyfftw.empty_aligned(

问题 I used to use fftw_plan_dft for multi-dimensional Fourier transformation. fftw_plan fftw_plan_dft(int rank, const int *n, fftw_complex *in, fftw_complex *out, int sign, unsigned flags); Now I want to pass 64 bit integer to fftw, it looks like I need to use fftw guru interface. fftw_plan fftw_plan_guru64_dft( int rank, const fftw_iodim64 *dims, int howmany_rank, const fftw_iodim64 *howmany_dims, fftw_complex *in, fftw_complex *out, int sign, unsigned flags); But I do not understand what is

问题 I am running my codes based on PyFFTW wrapper for FFTW3 library. In the plan function call of PyFFTW, there is a parameter "thread" to specify the number of threads to use. I use a fundamental judge, Activity Monitor, to tell whether multithreading works or not. It turns out that the following code runs on one CPU even when I specify thread=4. My laptop is a Mac with 8 cores. Here is my code. import pyfftw n_cpu = 4 flags = ('FFTW_MEASURE','FFTW_DESTROY_INPUT') x_buffer = pyfftw.n_byte_align

问题 I am trying to install pyFFTW on my Opensuse 13.1 x86_64 for a couple of days now and running into some issues all the time: I could not install pyFFTW for the first time due to some missing libraries of fftw. I found that it would be the best to install fftw from source, what I then did. This solves the problem of a missing library but a different problem arises, when I try to either build pyFFTW from source or by using pip for the installation: /usr/local/lib64/libfftw3.a: could not read