TF KB-NUFFT
GitHub |
Simple installation from pypi:
pip install tfkbnufft
About
This package is a verly early-stage and modest adaptation to TensorFlow of the torchkbnufft package written by Matthew Muckley for PyTorch. Please cite his work appropriately if you use this package.
Computation speed
The computation speeds are given in seconds, for a 256x256 image with a spokelength of 512 and 405 spokes. These numbers are not to be directly compared to those of torchkbnufft, since the computation is not the same. They are just to give a sense of the time required for computation.
| Operation | CPU | GPU |
|---|---|---|
| Forward NUFFT | 0.1676 | 0.0626 |
| Adjoint NUFFT | 0.7005 | 0.0635 |
To obtain these numbers for your machine, run the following commands, after installing this package:
pip install scikit-image Pillow
python profile_tfkbnufft.py
These numbers were obtained with a Quadro P5000.
Gradients
w.r.t trajectory
This is experimental currently and is WIP. Please be cautious. Currently this is tested in CI against results from NDFT, but clear mathematical backing to some aspects are still being understood for applying the chain rule.
References
-
Fessler, J. A., & Sutton, B. P. (2003). Nonuniform fast Fourier transforms using min-max interpolation. IEEE transactions on signal processing, 51(2), 560-574.
-
Beatty, P. J., Nishimura, D. G., & Pauly, J. M. (2005). Rapid gridding reconstruction with a minimal oversampling ratio. IEEE transactions on medical imaging, 24(6), 799-808.
-
Feichtinger, H. G., Gr, K., & Strohmer, T. (1995). Efficient numerical methods in non-uniform sampling theory. Numerische Mathematik, 69(4), 423-440.
Citation
If you want to cite the package, you can use any of the following:
@conference{muckley:20:tah,
author = {M. J. Muckley and R. Stern and T. Murrell and F. Knoll},
title = {{TorchKbNufft}: A High-Level, Hardware-Agnostic Non-Uniform Fast Fourier Transform},
booktitle = {ISMRM Workshop on Data Sampling \& Image Reconstruction},
year = 2020
}
@misc{Muckley2019,
author = {Muckley, M.J. et al.},
title = {Torch KB-NUFFT},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/mmuckley/torchkbnufft}}
}
Metadata
Release files for tfkbnufft 0.2.5
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| tfkbnufft-0.2.5.tar.gz | 21.9 kB | Without provenance | Details for tfkbnufft-0.2.5.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| tfkbnufft-0.2.5-py3-none-any.whl | Python 3 | none | any | Details for tfkbnufft-0.2.5-py3-none-any.whl |
Total release size: 46.3 kB
Release files / tfkbnufft-0.2.5.tar.gz
| Download URL | tfkbnufft-0.2.5.tar.gz |
|---|---|
| Size | 21.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
6e2c94424323185ed40d686906e4bbe0575f18fb4f18021dc046d7ee902377c4
|
|
BLAKE2b-256 checksum How to use checksums |
6bb53e3f9505d3620c7757d762e4fc472d2d2cff4ce3b8cb3f82746c77a51a04
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.11.2
|
Release files / tfkbnufft-0.2.5-py3-none-any.whl
| Download URL | tfkbnufft-0.2.5-py3-none-any.whl |
|---|---|
| Size | 24.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
131b13663f9bf7345093a5d1ac36ea63b44751e67abef3fc92f825942d32f4d4
|
|
BLAKE2b-256 checksum How to use checksums |
ba8c1248dc9d03b233ce0a0e3973948aa722f6e0cde7265747286c63623faeae
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.11.2
|
Security for tfkbnufft 0.2.5
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.