Python fft.fftn方法代码示例

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def run_lct(self, meas, X, Y, Z, S, x_max, y_max):
        self.max_dist = int(self.T * self.v/2 / self.channels[0])
        slope_x = self.dx * x_max / (self.fend/self.B * self.max_dist) * (1 + ((self.fstart/self.B)/(self.fend/self.B))**2)
        slope_y = self.dy * y_max / (self.fend/self.B * self.max_dist) * (1 + ((self.fstart/self.B)/(self.fend/self.B))**2)
        
        # params
        psf, fpsf = lct.getPSF(X, Y, Z, S, slope_x, slope_y)
        mtx, mtxi = lct.interpMtx(Z, S, self.fstart/self.B * self.max_dist, self.fend/self.B * self.max_dist)

        def pad_array(x, S, Z, X, Y):
            return np.pad(x, ((S*Z//2, S*Z//2), (Y//2, Y//2), (X//2, X//2)), 'constant')

        def trim_array(x, S, Z, X, Y):
            return x[S*int(np.floor(Z/2))+1:-S*int(np.ceil(Z/2))+1, Y//2+1:-Y//2+1, X//2+1:-X//2+1]

        invpsf = np.conj(fpsf) / (abs(fpsf)**2 + 1 / self.snr)
        tdata = np.matmul(mtx, meas.reshape((Z, -1))).reshape((-1, Y, X))

        fdata = fftn(pad_array(tdata, S, Z, X, Y))
        tvol = abs(trim_array(ifftn(fdata * invpsf), S, Z, X, Y))
        out = np.matmul(mtxi, tvol.reshape((S*Z, -1))).reshape((-1, Y, X))

        return out 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def getPSF(X, Y, Z, S, slope_x, slope_y):
    x = np.linspace(-1, 1, 2*X)
    y = np.linspace(-1, 1, 2*Y)
    z = np.linspace(0,2,2*S*Z)
    grid_z, grid_y, grid_x = np.meshgrid(z, y, x, indexing='ij')
    psf = np.abs(slope_x**2 * grid_x**2 + slope_y**2 * grid_y**2 - grid_z)

    psf = psf == np.tile(np.min(psf, axis=0, keepdims=True), (2*S*Z, 1, 1)) 
    psf = psf.astype(np.float32)
    psf = psf / np.sum(psf)

    psf = np.roll(psf, X, axis=2)
    psf = np.roll(psf, Y, axis=1)
    fpsf = fftn(psf)

    return psf, fpsf 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def lct(x1, y1, t1, v, vol, snr):
    X = len(x1)
    Y = len(y1)
    Z = len(t1)
    S = 2
    slope = np.max(x1) / (np.max(t1) * v/2)
    slope = np.max(y1) / (np.max(t1) * v/2)
    psf, fpsf = getPSF(X, Y, Z, S, slope)
    mtx, mtxi = interpMtx(Z, S, 0, np.max(t1)*v)

    def pad_array(x, S, Z, X):
        return np.pad(x, ((S*Z//2, S*Z//2), (X//2, X//2), (Y//2, Y//2)), 'constant')

    def trim_array(x, S, Z, X):
        return x[S*(Z//2)+1:-S*(Z//2)+1, X//2+1:-X//2+1, Y//2+1:-Y//2+1]

    invpsf = np.conj(fpsf) / (abs(fpsf)**2 + 1 / snr)
    tdata = np.matmul(mtx, vol)
    fdata = fftn(pad_array(tdata, S, Z, X))
    tvol = abs(trim_array(ifftn(fdata * invpsf), S, Z, X))
    vol = np.matmul(mtxi, tvol)
    return vol 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def fl2norm2(xf, axis=(0, 1)):
    r"""Compute the squared :math:`\ell_2` norm in the DFT domain.

    Compute the squared :math:`\ell_2` norm in the DFT domain, taking
    into account the unnormalised DFT scaling, i.e. given the DFT of a
    multi-dimensional array computed via :func:`fftn`, return the
    squared :math:`\ell_2` norm of the original array.

    Parameters
    ----------
    xf : array_like
      Input array
    axis : sequence of ints, optional (default (0,1))
      Axes on which the input is in the frequency domain

    Returns
    -------
    x : float
      :math:`\|\mathbf{x}\|_2^2` where the input array is the result of
      applying :func:`fftn` to the specified axes of multi-dimensional
      array :math:`\mathbf{x}`
    """

    xfs = xf.shape
    return (np.linalg.norm(xf)**2) / np.prod(np.array([xfs[k] for k in axis])) 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def run_test_c2c_impl(self, shape, axes, inverse=False, fftshift=False):
        shape = list(shape)
        shape[-1] *= 2 # For complex
        known_data = np.random.normal(size=shape).astype(np.float32).view(np.complex64)
        idata = bf.ndarray(known_data, space='cuda')
        odata = bf.empty_like(idata)
        fft = Fft()
        fft.init(idata, odata, axes=axes, apply_fftshift=fftshift)
        fft.execute(idata, odata, inverse)
        if inverse:
            if fftshift:
                known_data = np.fft.ifftshift(known_data, axes=axes)
            # Note: Numpy applies normalization while CUFFT does not
            norm = reduce(lambda a, b: a * b, [known_data.shape[d]
                                               for d in axes])
            known_result = gold_ifftn(known_data, axes=axes) * norm
        else:
            known_result = gold_fftn(known_data, axes=axes)
            if fftshift:
                known_result = np.fft.fftshift(known_result, axes=axes)
        x = (np.abs(odata.copy('system') - known_result) / known_result > RTOL).astype(np.int32)
        a = odata.copy('system')
        b = known_result
        compare(odata.copy('system'), known_result) 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def get_numpy(shape, fftn_shape=None, **kwargs):
    import numpy.fft as numpy_fft

    f = {
        "fft2": numpy_fft.fft2,
        "ifft2": numpy_fft.ifft2,
        "rfft2": numpy_fft.rfft2,
        "irfft2": lambda X: numpy_fft.irfft2(X, s=shape),
        "fftshift": numpy_fft.fftshift,
        "ifftshift": numpy_fft.ifftshift,
        "fftfreq": numpy_fft.fftfreq,
    }
    if fftn_shape is not None:
        f["fftn"] = numpy_fft.fftn
    fft = SimpleNamespace(**f)

    return fft 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def get_scipy(shape, fftn_shape=None, **kwargs):
    import numpy.fft as numpy_fft
    import scipy.fftpack as scipy_fft

    # use numpy implementation of rfft2/irfft2 because they have not been
    # implemented in scipy.fftpack
    f = {
        "fft2": scipy_fft.fft2,
        "ifft2": scipy_fft.ifft2,
        "rfft2": numpy_fft.rfft2,
        "irfft2": lambda X: numpy_fft.irfft2(X, s=shape),
        "fftshift": scipy_fft.fftshift,
        "ifftshift": scipy_fft.ifftshift,
        "fftfreq": scipy_fft.fftfreq,
    }
    if fftn_shape is not None:
        f["fftn"] = scipy_fft.fftn
    fft = SimpleNamespace(**f)

    return fft 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def potential(self, q, steps):
        hx = steps[0]
        hy = steps[1]
        hz = steps[2]
        Nx = q.shape[0]
        Ny = q.shape[1]
        Nz = q.shape[2]
        out = np.zeros((2*Nx-1, 2*Ny-1, 2*Nz-1))
        out[:Nx, :Ny, :Nz] = q
        K1 = self.sym_kernel(q.shape, steps)
        K2 = np.zeros((2*Nx-1, 2*Ny-1, 2*Nz-1))
        K2[0:Nx, 0:Ny, 0:Nz] = K1
        K2[0:Nx, 0:Ny, Nz:2*Nz-1] = K2[0:Nx, 0:Ny, Nz-1:0:-1] #z-mirror
        K2[0:Nx, Ny:2*Ny-1,:] = K2[0:Nx, Ny-1:0:-1, :]        #y-mirror
        K2[Nx:2*Nx-1, :, :] = K2[Nx-1:0:-1, :, :]             #x-mirror
        t0 = time.time()
        if pyfftw_flag:
            nthreads = int(conf.OCELOT_NUM_THREADS)
            if nthreads < 1:
                nthreads = 1
            K2_fft = pyfftw.builders.fftn(K2, axes=None, overwrite_input=False, planner_effort='FFTW_ESTIMATE',
                                       threads=nthreads, auto_align_input=False, auto_contiguous=False, avoid_copy=True)
            out_fft = pyfftw.builders.fftn(out, axes=None, overwrite_input=False, planner_effort='FFTW_ESTIMATE',
                                          threads=nthreads, auto_align_input=False, auto_contiguous=False, avoid_copy=True)
            out_ifft = pyfftw.builders.ifftn(out_fft()*K2_fft(), axes=None, overwrite_input=False, planner_effort='FFTW_ESTIMATE',
                                          threads=nthreads, auto_align_input=False, auto_contiguous=False, avoid_copy=True)
            out = np.real(out_ifft())

        else:
            out = np.real(ifftn(fftn(out)*fftn(K2)))
        t1 = time.time()
        logger.debug('fft time:' + str(t1-t0) + ' sec')
        out[:Nx, :Ny, :Nz] = out[:Nx,:Ny,:Nz]/(4*pi*epsilon_0*hx*hy*hz)
        return out[:Nx, :Ny, :Nz] 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def bench_random(self):
        from numpy.fft import fftn as numpy_fftn
        print()
        print('    Multi-dimensional Fast Fourier Transform')
        print('===================================================')
        print('          |    real input     |   complex input    ')
        print('---------------------------------------------------')
        print('   size   |  scipy  |  numpy  |  scipy  |  numpy ')
        print('---------------------------------------------------')
        for size,repeat in [((100,100),100),((1000,100),7),
                            ((256,256),10),
                            ((512,512),3),
                            ]:
            print('%9s' % ('%sx%s' % size), end=' ')
            sys.stdout.flush()

            for x in [random(size).astype(double),
                      random(size).astype(cdouble)+random(size).astype(cdouble)*1j
                      ]:
                y = fftn(x)
                #if size > 500: y = fftn(x)
                #else: y = direct_dft(x)
                assert_array_almost_equal(fftn(x),y)
                print('|%8.2f' % measure('fftn(x)',repeat), end=' ')
                sys.stdout.flush()

                assert_array_almost_equal(numpy_fftn(x),y)
                print('|%8.2f' % measure('numpy_fftn(x)',repeat), end=' ')
                sys.stdout.flush()

            print(' (secs for %s calls)' % (repeat))

        sys.stdout.flush() 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def fftd(I, dims=None):

    # Compute fft
    if dims is None:
        X = fftn(I)
    elif dims == 2:
        X = fft2(I, axes=(0, 1))
    else:
        X = fftn(I, axes=tuple(range(dims)))

    return X 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def fftnc(x, axes, ortho=True):
    tmp = fft.fftshift(x, axes=axes)
    tmp = fft.fftn(tmp, axes=axes, norm="ortho" if ortho else None)
    return fft.ifftshift(tmp, axes=axes) 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def filter_given_curve(v, curve):
    grid = GV.grid_displacement_to_center(v.shape, GV.fft_mid_co(v.shape))
    rad = GV.grid_distance_to_center(grid)
    rad = N.round(rad).astype(N.int)
    b = N.zeros(rad.shape)
    for (i, a) in enumerate(curve):
        b[(rad == i)] = a
    vf = ifftn(ifftshift((fftshift(fftn(v)) * b)))
    vf = N.real(vf)
    return vf 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def __init__(self, images, band_width_radius=1.0):

        im_f = {}
        for k in range(len(images)):
            im = images[k]
            im = fftshift(fftn(im))
            im_f[k] = im

        self.im_f = im_f        # fft transformed images
        self.ks = set()
        self.img_siz = im_f[k].shape
        self.set_fft_mid_co()
        self.set_rad()
        self.band_width_radius = band_width_radius 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def __init__(self, images, masks, band_width_radius=1.0):

        im_f = {}
        for k in images:
            im = images[k]
            im = fftshift(fftn(im))
            im_f[k] = im

        self.im_f = im_f        # fft transformed images
        self.ms = masks         # masks
        self.ks = set()
        self.img_siz = im_f[k].shape
        self.set_fft_mid_co()
        self.set_rad()
        self.band_width_radius = band_width_radius 

# 需要导入模块: from numpy import fft [as 别名]
# 或者: from numpy.fft import fftn [as 别名]
def fourier_transform(v):
    return fftshift(fftn(v))