Skip to content

replace fwd_scale with norm kwarg in mkl_fft #189

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: revisit_overwrite_x
Choose a base branch
from
Open

replace fwd_scale with norm kwarg in mkl_fft #189

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@ All notable changes to this project will be documented in this file.
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## [2.0.0] (05/DD/2025)
## [2.0.0] (06/DD/2025)

### Added
* Added Hermitian FFT functions to SciPy interface `mkl_fft.interfaces.scipy_fft`: `hfft`, `ihfft`, `hfftn`, `ihfftn`, `hfft2`, and `ihfft2` [gh-161](https://github.com/IntelPython/mkl_fft/pull/161)
Expand All @@ -17,6 +17,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
* SciPy interface `mkl_fft.interfaces.scipy_fft` uses the same function from SciPy for handling `s` and `axes` for N-D FFTs [gh-181](https://github.com/IntelPython/mkl_fft/pull/181)
* Dropped support for `scipy.fftpack` interface [gh-185](https://github.com/IntelPython/mkl_fft/pull/185)
* Dropped support for `overwrite_x` parameter in `mkl_fft` [gh-185](https://github.com/IntelPython/mkl_fft/pull/185)
* Replaced `fwd_scale` parameter with `norm` in `mkl_fft` [gh-189](https://github.com/IntelPython/mkl_fft/pull/189)

### Fixed
* Fixed a bug in `mkl_fft.interfaces.numpy.fftn` when an empty tuple is passed for `axes` [gh-139](https://github.com/IntelPython/mkl_fft/pull/139)
Expand Down
12 changes: 6 additions & 6 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,21 +52,21 @@ While using the interfaces module is the recommended way to leverage `mk_fft`, o

### complex-to-complex (c2c) transforms:

`fft(x, n=None, axis=-1, fwd_scale=1.0, out=None)` - 1D FFT, similar to `numpy.fft.fft`
`fft(x, n=None, axis=-1, norm=None, out=None)` - 1D FFT, similar to `numpy.fft.fft`

`fft2(x, s=None, axes=(-2, -1), fwd_scale=1.0, out=None)` - 2D FFT, similar to `numpy.fft.fft2`
`fft2(x, s=None, axes=(-2, -1), norm=None, out=None)` - 2D FFT, similar to `numpy.fft.fft2`

`fftn(x, s=None, axes=None, fwd_scale=1.0, out=None)` - ND FFT, similar to `numpy.fft.fftn`
`fftn(x, s=None, axes=None, norm=None, out=None)` - ND FFT, similar to `numpy.fft.fftn`

and similar inverse FFT (`ifft*`) functions.

### real-to-complex (r2c) and complex-to-real (c2r) transforms:

`rfft(x, n=None, axis=-1, fwd_scale=1.0, out=None)` - r2c 1D FFT, similar to `numpy.fft.rfft`
`rfft(x, n=None, axis=-1, norm=None, out=None)` - r2c 1D FFT, similar to `numpy.fft.rfft`

`rfft2(x, s=None, axes=(-2, -1), fwd_scale=1.0, out=None)` - r2c 2D FFT, similar to `numpy.fft.rfft2`
`rfft2(x, s=None, axes=(-2, -1), norm=None, out=None)` - r2c 2D FFT, similar to `numpy.fft.rfft2`

`rfftn(x, s=None, axes=None, fwd_scale=1.0, out=None)` - r2c ND FFT, similar to `numpy.fft.rfftn`
`rfftn(x, s=None, axes=None, norm=None, out=None)` - r2c ND FFT, similar to `numpy.fft.rfftn`

and similar inverse c2r FFT (`irfft*`) functions.

Expand Down
12 changes: 4 additions & 8 deletions mkl_fft/_fft_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -198,16 +198,12 @@ def _init_nd_shape_and_axes(x, shape, axes):
raise ValueError("when given, shape values must be integers")
if axes.shape != shape.shape:
raise ValueError(
"when given, axes and shape arguments"
" have to be of the same length"
"when given, axes and shape arguments have to be of the same length"
)

shape = np.where(shape == -1, np.array(x.shape)[axes], shape)

if shape.size != 0 and (shape < 1).any():
raise ValueError(
"invalid number of data points ({0}) specified".format(shape)
)
raise ValueError(f"invalid number of data points ({shape}) specified")

return shape, axes

Expand Down Expand Up @@ -319,7 +315,7 @@ def _pad_array(arr, s, axes):
try:
shp_i = arr_shape[ai]
except IndexError:
raise ValueError("Invalid axis (%d) specified" % ai)
raise ValueError(f"Invalid axis {ai} specified")
if si > shp_i:
no_padding = False
pad_widths[ai] = (0, si - shp_i)
Expand Down Expand Up @@ -355,7 +351,7 @@ def _trim_array(arr, s, axes):
try:
shp_i = arr_shape[ai]
except IndexError:
raise ValueError("Invalid axis (%d) specified" % ai)
raise ValueError(f"Invalid axis {ai} specified")
if si < shp_i:
no_trim = False
ind[ai] = slice(None, si, None)
Expand Down
71 changes: 38 additions & 33 deletions mkl_fft/_mkl_fft.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,12 @@
# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

from ._fft_utils import _c2c_fftnd_impl, _c2r_fftnd_impl, _r2c_fftnd_impl
from ._fft_utils import (
_c2c_fftnd_impl,
_c2r_fftnd_impl,
_compute_fwd_scale,
_r2c_fftnd_impl,
)

# pylint: disable=no-name-in-module
from ._pydfti import _c2c_fft1d_impl, _c2r_fft1d_impl, _r2c_fft1d_impl
Expand All @@ -45,57 +50,57 @@
]


def fft(x, n=None, axis=-1, fwd_scale=1.0, out=None):
return _c2c_fft1d_impl(
x, n=n, axis=axis, out=out, direction=+1, fsc=fwd_scale
)
def fft(x, n=None, axis=-1, norm=None, out=None):
fsc = _compute_fwd_scale(norm, n, x.shape[axis])
return _c2c_fft1d_impl(x, n=n, axis=axis, out=out, direction=+1, fsc=fsc)


def ifft(x, n=None, axis=-1, fwd_scale=1.0, out=None):
return _c2c_fft1d_impl(
x, n=n, axis=axis, out=out, direction=-1, fsc=fwd_scale
)
def ifft(x, n=None, axis=-1, norm=None, out=None):
fsc = _compute_fwd_scale(norm, n, x.shape[axis])
return _c2c_fft1d_impl(x, n=n, axis=axis, out=out, direction=-1, fsc=fsc)


def fft2(x, s=None, axes=(-2, -1), fwd_scale=1.0, out=None):
return fftn(x, s=s, axes=axes, out=out, fwd_scale=fwd_scale)
def fft2(x, s=None, axes=(-2, -1), norm=None, out=None):
return fftn(x, s=s, axes=axes, norm=norm, out=out)


def ifft2(x, s=None, axes=(-2, -1), fwd_scale=1.0, out=None):
return ifftn(x, s=s, axes=axes, out=out, fwd_scale=fwd_scale)
def ifft2(x, s=None, axes=(-2, -1), norm=None, out=None):
return ifftn(x, s=s, axes=axes, norm=norm, out=out)


def fftn(x, s=None, axes=None, fwd_scale=1.0, out=None):
return _c2c_fftnd_impl(
x, s=s, axes=axes, out=out, direction=+1, fsc=fwd_scale
)
def fftn(x, s=None, axes=None, norm=None, out=None):
fsc = _compute_fwd_scale(norm, s, x.shape)
return _c2c_fftnd_impl(x, s=s, axes=axes, out=out, direction=+1, fsc=fsc)


def ifftn(x, s=None, axes=None, fwd_scale=1.0, out=None):
return _c2c_fftnd_impl(
x, s=s, axes=axes, out=out, direction=-1, fsc=fwd_scale
)
def ifftn(x, s=None, axes=None, norm=None, out=None):
fsc = _compute_fwd_scale(norm, s, x.shape)
return _c2c_fftnd_impl(x, s=s, axes=axes, out=out, direction=-1, fsc=fsc)


def rfft(x, n=None, axis=-1, fwd_scale=1.0, out=None):
return _r2c_fft1d_impl(x, n=n, axis=axis, out=out, fsc=fwd_scale)
def rfft(x, n=None, axis=-1, norm=None, out=None):
fsc = _compute_fwd_scale(norm, n, x.shape[axis])
return _r2c_fft1d_impl(x, n=n, axis=axis, out=out, fsc=fsc)


def irfft(x, n=None, axis=-1, fwd_scale=1.0, out=None):
return _c2r_fft1d_impl(x, n=n, axis=axis, out=out, fsc=fwd_scale)
def irfft(x, n=None, axis=-1, norm=None, out=None):
fsc = _compute_fwd_scale(norm, n, 2 * (x.shape[axis] - 1))
return _c2r_fft1d_impl(x, n=n, axis=axis, out=out, fsc=fsc)


def rfft2(x, s=None, axes=(-2, -1), fwd_scale=1.0, out=None):
return rfftn(x, s=s, axes=axes, out=out, fwd_scale=fwd_scale)
def rfft2(x, s=None, axes=(-2, -1), norm=None, out=None):
return rfftn(x, s=s, axes=axes, norm=norm, out=out)


def irfft2(x, s=None, axes=(-2, -1), fwd_scale=1.0, out=None):
return irfftn(x, s=s, axes=axes, out=out, fwd_scale=fwd_scale)
def irfft2(x, s=None, axes=(-2, -1), norm=None, out=None):
return irfftn(x, s=s, axes=axes, norm=norm, out=out)


def rfftn(x, s=None, axes=None, fwd_scale=1.0, out=None):
return _r2c_fftnd_impl(x, s=s, axes=axes, out=out, fsc=fwd_scale)
def rfftn(x, s=None, axes=None, norm=None, out=None):
fsc = _compute_fwd_scale(norm, s, x.shape)
return _r2c_fftnd_impl(x, s=s, axes=axes, out=out, fsc=fsc)


def irfftn(x, s=None, axes=None, fwd_scale=1.0, out=None):
return _c2r_fftnd_impl(x, s=s, axes=axes, out=out, fsc=fwd_scale)
def irfftn(x, s=None, axes=None, norm=None, out=None):
fsc = _compute_fwd_scale(norm, s, x.shape)
return _c2r_fftnd_impl(x, s=s, axes=axes, out=out, fsc=fsc)
33 changes: 11 additions & 22 deletions mkl_fft/interfaces/_numpy_fft.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,7 @@

import mkl_fft

from .._fft_utils import _compute_fwd_scale, _swap_direction
from .._fft_utils import _swap_direction
from ._float_utils import _downcast_float128_array

__all__ = [
Expand Down Expand Up @@ -120,10 +120,9 @@ def fft(a, n=None, axis=-1, norm=None, out=None):

"""
x = _downcast_float128_array(a)
fsc = _compute_fwd_scale(norm, n, x.shape[axis])

return _trycall(
mkl_fft.fft, (x,), {"n": n, "axis": axis, "fwd_scale": fsc, "out": out}
mkl_fft.fft, (x,), {"n": n, "axis": axis, "norm": norm, "out": out}
)


Expand All @@ -135,10 +134,9 @@ def ifft(a, n=None, axis=-1, norm=None, out=None):

"""
x = _downcast_float128_array(a)
fsc = _compute_fwd_scale(norm, n, x.shape[axis])

return _trycall(
mkl_fft.ifft, (x,), {"n": n, "axis": axis, "fwd_scale": fsc, "out": out}
mkl_fft.ifft, (x,), {"n": n, "axis": axis, "norm": norm, "out": out}
)


Expand Down Expand Up @@ -171,10 +169,9 @@ def fftn(a, s=None, axes=None, norm=None, out=None):
"""
x = _downcast_float128_array(a)
s, axes = _cook_nd_args(x, s, axes)
fsc = _compute_fwd_scale(norm, s, x.shape)

return _trycall(
mkl_fft.fftn, (x,), {"s": s, "axes": axes, "fwd_scale": fsc, "out": out}
mkl_fft.fftn, (x,), {"s": s, "axes": axes, "norm": norm, "out": out}
)


Expand All @@ -187,12 +184,11 @@ def ifftn(a, s=None, axes=None, norm=None, out=None):
"""
x = _downcast_float128_array(a)
s, axes = _cook_nd_args(x, s, axes)
fsc = _compute_fwd_scale(norm, s, x.shape)

return _trycall(
mkl_fft.ifftn,
(x,),
{"s": s, "axes": axes, "fwd_scale": fsc, "out": out},
{"s": s, "axes": axes, "norm": norm, "out": out},
)


Expand All @@ -204,10 +200,9 @@ def rfft(a, n=None, axis=-1, norm=None, out=None):

"""
x = _downcast_float128_array(a)
fsc = _compute_fwd_scale(norm, n, x.shape[axis])

return _trycall(
mkl_fft.rfft, (x,), {"n": n, "axis": axis, "fwd_scale": fsc, "out": out}
mkl_fft.rfft, (x,), {"n": n, "axis": axis, "norm": norm, "out": out}
)


Expand All @@ -219,12 +214,11 @@ def irfft(a, n=None, axis=-1, norm=None, out=None):

"""
x = _downcast_float128_array(a)
fsc = _compute_fwd_scale(norm, n, 2 * (x.shape[axis] - 1))

return _trycall(
mkl_fft.irfft,
(x,),
{"n": n, "axis": axis, "fwd_scale": fsc, "out": out},
{"n": n, "axis": axis, "norm": norm, "out": out},
)


Expand Down Expand Up @@ -257,12 +251,11 @@ def rfftn(a, s=None, axes=None, norm=None, out=None):
"""
x = _downcast_float128_array(a)
s, axes = _cook_nd_args(x, s, axes)
fsc = _compute_fwd_scale(norm, s, x.shape)

return _trycall(
mkl_fft.rfftn,
(x,),
{"s": s, "axes": axes, "fwd_scale": fsc, "out": out},
{"s": s, "axes": axes, "norm": norm, "out": out},
)


Expand All @@ -276,12 +269,11 @@ def irfftn(a, s=None, axes=None, norm=None, out=None):

x = _downcast_float128_array(a)
s, axes = _cook_nd_args(x, s, axes, invreal=True)
fsc = _compute_fwd_scale(norm, s, x.shape)

return _trycall(
mkl_fft.irfftn,
(x,),
{"s": s, "axes": axes, "fwd_scale": fsc, "out": out},
{"s": s, "axes": axes, "norm": norm, "out": out},
)


Expand All @@ -295,12 +287,10 @@ def hfft(a, n=None, axis=-1, norm=None, out=None):
"""
norm = _swap_direction(norm)
x = _downcast_float128_array(a)
fsc = _compute_fwd_scale(norm, n, 2 * (x.shape[axis] - 1))

return _trycall(
mkl_fft.irfft,
(np.conjugate(x),),
{"n": n, "axis": axis, "fwd_scale": fsc, "out": out},
{"n": n, "axis": axis, "norm": norm, "out": out},
)


Expand All @@ -313,10 +303,9 @@ def ihfft(a, n=None, axis=-1, norm=None, out=None):
"""
norm = _swap_direction(norm)
x = _downcast_float128_array(a)
fsc = _compute_fwd_scale(norm, n, x.shape[axis])

result = _trycall(
mkl_fft.rfft, (x,), {"n": n, "axis": axis, "fwd_scale": fsc, "out": out}
mkl_fft.rfft, (x,), {"n": n, "axis": axis, "norm": norm, "out": out}
)

np.conjugate(result, out=result)
Expand Down
Loading
Loading