## ``mkl_fft`` -- a NumPy-based Python interface to Intel (R) MKL FFT functionality
[](https://travis-ci.com/IntelPython/mkl_fft)
`mkl_fft` started as a part of Intel (R) Distribution for Python* optimizations to NumPy, and is now being released
as a stand-alone package. It can be installed into conda environment using
```
conda install -c intel mkl_fft
```
---
To install mkl_fft Pypi package please use following command:
```
python -m pip install --index-url https://pypi.anaconda.org/intel/simple --extra-index-url https://pypi.org/simple mkl_fft
```
If command above installs NumPy package from the Pypi, please use following command to install Intel optimized NumPy wheel package from Anaconda Cloud:
```
python -m pip install --index-url https://pypi.anaconda.org/intel/simple --extra-index-url https://pypi.org/simple mkl_fft numpy==<numpy_version>
```
Where `<numpy_version>` should be the latest version from https://anaconda.org/intel/numpy
---
Since MKL FFT supports performing discrete Fourier transforms over non-contiguously laid out arrays, MKL can be directly
used on any well-behaved floating point array with no internal overlaps for both in-place and not in-place transforms of
arrays in single and double floating point precision.
This eliminates the need to copy input array contiguously into an intermediate buffer.
`mkl_fft` directly supports N-dimensional Fourier transforms.
More details can be found in SciPy 2017 conference proceedings:
https://github.com/scipy-conference/scipy_proceedings/tree/2017/papers/oleksandr_pavlyk
---
It implements the following functions:
### Complex transforms, similar to those in `scipy.fftpack`:
`fft(x, n=None, axis=-1, overwrite_x=False)`
`ifft(x, n=None, axis=-1, overwrite_x=False)`
`fft2(x, shape=None, axes=(-2,-1), overwrite_x=False)`
`ifft2(x, shape=None, axes=(-2,-1), overwrite_x=False)`
`fftn(x, n=None, axes=None, overwrite_x=False)`
`ifftn(x, n=None, axes=None, overwrite_x=False)`
### Real transforms
`rfft(x, n=None, axis=-1, overwrite_x=False)` - real 1D Fourier transform, like `scipy.fftpack.rfft`
`rfft_numpy(x, n=None, axis=-1)` - real 1D Fourier transform, like `numpy.fft.rfft`
`rfft2_numpy(x, s=None, axes=(-2,-1))` - real 2D Fourier transform, like `numpy.fft.rfft2`
`rfftn_numpy(x, s=None, axes=None)` - real 2D Fourier transform, like `numpy.fft.rfftn`
... and similar `irfft*` functions.
The package also provides `mkl_fft._numpy_fft` and `mkl_fft._scipy_fft` interfaces which provide drop-in replacements for equivalent functions in NumPy and SciPy respectively.
---
To build ``mkl_fft`` from sources on Linux:
- install a recent version of MKL, if necessary;
- execute ``source /path/to/mklroot/bin/mklvars.sh intel64`` ;
- execute ``pip install .``
Raw data
{
"_id": null,
"home_page": "http://github.com/IntelPython/mkl_fft",
"name": "mkl-fft",
"maintainer": "Intel Corp.",
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": "scripting@intel.com",
"keywords": "DFTI, FFT, Fourier, MKL",
"author": "Intel Corporation",
"author_email": null,
"download_url": "http://github.com/IntelPython/mkl_fft",
"platform": "Windows",
"description": "## ``mkl_fft`` -- a NumPy-based Python interface to Intel (R) MKL FFT functionality\n[](https://travis-ci.com/IntelPython/mkl_fft)\n\n`mkl_fft` started as a part of Intel (R) Distribution for Python* optimizations to NumPy, and is now being released\nas a stand-alone package. It can be installed into conda environment using\n\n```\n conda install -c intel mkl_fft\n```\n\n---\n\nTo install mkl_fft Pypi package please use following command:\n\n```\n python -m pip install --index-url https://pypi.anaconda.org/intel/simple --extra-index-url https://pypi.org/simple mkl_fft\n```\n\nIf command above installs NumPy package from the Pypi, please use following command to install Intel optimized NumPy wheel package from Anaconda Cloud:\n\n```\n python -m pip install --index-url https://pypi.anaconda.org/intel/simple --extra-index-url https://pypi.org/simple mkl_fft numpy==<numpy_version>\n```\n\nWhere `<numpy_version>` should be the latest version from https://anaconda.org/intel/numpy\n\n---\n\nSince MKL FFT supports performing discrete Fourier transforms over non-contiguously laid out arrays, MKL can be directly\nused on any well-behaved floating point array with no internal overlaps for both in-place and not in-place transforms of\narrays in single and double floating point precision.\n\nThis eliminates the need to copy input array contiguously into an intermediate buffer.\n\n`mkl_fft` directly supports N-dimensional Fourier transforms.\n\nMore details can be found in SciPy 2017 conference proceedings:\n https://github.com/scipy-conference/scipy_proceedings/tree/2017/papers/oleksandr_pavlyk\n\n---\n\nIt implements the following functions:\n\n### Complex transforms, similar to those in `scipy.fftpack`:\n\n`fft(x, n=None, axis=-1, overwrite_x=False)`\n\n`ifft(x, n=None, axis=-1, overwrite_x=False)`\n\n`fft2(x, shape=None, axes=(-2,-1), overwrite_x=False)`\n\n`ifft2(x, shape=None, axes=(-2,-1), overwrite_x=False)`\n\n`fftn(x, n=None, axes=None, overwrite_x=False)`\n\n`ifftn(x, n=None, axes=None, overwrite_x=False)`\n\n### Real transforms\n\n`rfft(x, n=None, axis=-1, overwrite_x=False)` - real 1D Fourier transform, like `scipy.fftpack.rfft`\n\n`rfft_numpy(x, n=None, axis=-1)` - real 1D Fourier transform, like `numpy.fft.rfft`\n\n`rfft2_numpy(x, s=None, axes=(-2,-1))` - real 2D Fourier transform, like `numpy.fft.rfft2`\n\n`rfftn_numpy(x, s=None, axes=None)` - real 2D Fourier transform, like `numpy.fft.rfftn`\n\n... and similar `irfft*` functions.\n\n\nThe package also provides `mkl_fft._numpy_fft` and `mkl_fft._scipy_fft` interfaces which provide drop-in replacements for equivalent functions in NumPy and SciPy respectively.\n\n---\n\nTo build ``mkl_fft`` from sources on Linux:\n - install a recent version of MKL, if necessary;\n - execute ``source /path/to/mklroot/bin/mklvars.sh intel64`` ;\n - execute ``pip install .``\n\n\n",
"bugtrack_url": null,
"license": "BSD",
"summary": "MKL-based FFT transforms for NumPy arrays",
"version": "1.3.8",
"project_urls": {
"Download": "http://github.com/IntelPython/mkl_fft",
"Homepage": "http://github.com/IntelPython/mkl_fft"
},
"split_keywords": [
"dfti",
" fft",
" fourier",
" mkl"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "f5206eef8c59d63043da6a8339be7afb60f9b9ead43470efe28e29b38ff9e38b",
"md5": "dbe3b53d6da1e90044d7247fba11c5c7",
"sha256": "1640b75a5d47bb99c944815f0e349e9f27d9a107ae1c83a4ea3a7726f118a085"
},
"downloads": -1,
"filename": "mkl_fft-1.3.8-70-cp310-cp310-manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "dbe3b53d6da1e90044d7247fba11c5c7",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.7",
"size": 4079290,
"upload_time": "2024-03-26T22:33:36",
"upload_time_iso_8601": "2024-03-26T22:33:36.087722Z",
"url": "https://files.pythonhosted.org/packages/f5/20/6eef8c59d63043da6a8339be7afb60f9b9ead43470efe28e29b38ff9e38b/mkl_fft-1.3.8-70-cp310-cp310-manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "e839caa373695ed95a9f78e99f53c4a4fbff86784f56161b0a4b0abc93e31b79",
"md5": "d027632bbb543238de8b87ef8ff760f5",
"sha256": "32260b4f802fc5185ead67a032e844809638cfd44c1b4b76e908cd74d2f84926"
},
"downloads": -1,
"filename": "mkl_fft-1.3.8-70-cp310-cp310-win_amd64.whl",
"has_sig": false,
"md5_digest": "d027632bbb543238de8b87ef8ff760f5",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.7",
"size": 174558,
"upload_time": "2024-03-26T22:33:39",
"upload_time_iso_8601": "2024-03-26T22:33:39.264173Z",
"url": "https://files.pythonhosted.org/packages/e8/39/caa373695ed95a9f78e99f53c4a4fbff86784f56161b0a4b0abc93e31b79/mkl_fft-1.3.8-70-cp310-cp310-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "e184d5da4cbaea6e9961ac81e84e769213e84af105d5c5a04061c232ee0d577b",
"md5": "e40fd148ed2b7982fa1b67d7b172cec2",
"sha256": "2b77d42a808d7e789eae0f5a5436c37e7979c24ffe83a559c8c62c591d00b7a5"
},
"downloads": -1,
"filename": "mkl_fft-1.3.8-70-cp39-cp39-manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "e40fd148ed2b7982fa1b67d7b172cec2",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.7",
"size": 4084054,
"upload_time": "2024-03-26T22:33:31",
"upload_time_iso_8601": "2024-03-26T22:33:31.013869Z",
"url": "https://files.pythonhosted.org/packages/e1/84/d5da4cbaea6e9961ac81e84e769213e84af105d5c5a04061c232ee0d577b/mkl_fft-1.3.8-70-cp39-cp39-manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "6e7123a25b0bddef29eb84a7e948d1092618894e4211ef937f61f4a7444153f2",
"md5": "0985d53e8405364c1147804bd188baa2",
"sha256": "01bc0c9df42c18d8e25f633497b9ecf81088d8dadd39c7917fe105d3e219c124"
},
"downloads": -1,
"filename": "mkl_fft-1.3.8-70-cp39-cp39-win_amd64.whl",
"has_sig": false,
"md5_digest": "0985d53e8405364c1147804bd188baa2",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.7",
"size": 178000,
"upload_time": "2024-03-26T22:33:33",
"upload_time_iso_8601": "2024-03-26T22:33:33.602767Z",
"url": "https://files.pythonhosted.org/packages/6e/71/23a25b0bddef29eb84a7e948d1092618894e4211ef937f61f4a7444153f2/mkl_fft-1.3.8-70-cp39-cp39-win_amd64.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-03-26 22:33:36",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "IntelPython",
"github_project": "mkl_fft",
"travis_ci": true,
"coveralls": false,
"github_actions": true,
"lcname": "mkl-fft"
}
JSON
