This is a fork of the Gpufit library that is modified directly for use with Peter Spencer's Interferometry Project.
His project provides the necessary modified files, with the expectation that you simply swap the files into GpuFit and build it from source. Here, the files are already in place and ready to build. Additionally, the repository is frozen at an older version, since peter's code has been not been touched in a while.
This repository has also been significantly chopped down; CPUFit, tests, Java, etc have all been removed.
Most of the original Readme has been left in tact below. You will have to compile from source. The following works for me, on Arch Linux
git clone https://github.com/andresh3/Gpufit-Custom.git
mkdir build
cd build
cmake -DCMAKE_BUILD_TYPE=RELEASE -DCUDA_ARCHITECTURES=Auto ../Gpufit-Custom/
make
Once you've built from source, you can install the Gpufit python package
CD into your project directory, and optionally activate your virtual environment:
cd /Path/To/Project
source venv/bin/activate
Then install the wheel (whatever the .whl file in dist is called use it):
pip install /Path/To/build/Gpufit/pyGpufit/dist/pygpufit-1.2.0-py2.py3-none-any.whl
Levenberg Marquardt curve fitting in CUDA.
Homepage: github.com/gpufit/Gpufit
The manuscript describing Gpufit is now published in Scientific Reports.
To verify that Gpufit is working correctly on the host computer, go to the folder gpufit_performance_test of the binary package and run Gpufit_Cpufit_Performance_Comparison.exe. Further details of the test executable can be found in the documentation package.
The latest Gpufit binary release, supporting Windows 32-bit and 64-bit machines, can be found on the release page.
Documentation for the Gpufit library may be found online (latest documentation), and also as a PDF file in the binary distribution of Gpufit.
Instructions for building Gpufit are found in the documentation: Building from source code.
Gpufit was created by Mark Bates, Adrian Przybylski, Björn Thiel, and Jan Keller-Findeisen at the Max Planck Institute for Biophysical Chemistry, in Göttingen, Germany.
This specific repository has code from Peter Spencer (see repo at top), and has been forked/assembled by Andrew Humphreys.
If you use Gpufit in your research, please cite our publication describing the software. A paper describing the software was published in Scientific Reports. The open-access manuscript is available from the Scientific Reports website, here.
- Gpufit: An open-source toolkit for GPU-accelerated curve fitting
Adrian Przybylski, Björn Thiel, Jan Keller-Findeisen, Bernd Stock, and Mark Bates
Scientific Reports, vol. 7, 15722 (2017); doi: https://doi.org/10.1038/s41598-017-15313-9
MIT License
Copyright (c) 2017 Mark Bates, Adrian Przybylski, Björn Thiel, and Jan Keller-Findeisen
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.