This package contains analysis code used by the Stanford optical levitation lab to search for non-Newtonian interactions at the micron scale.
It is highly recommended that you install this inside a virtual environment using venv or conda, since the specific numpy and scipy versions listed in requirements.txt are required in order to avoid known performance issues with later versions. From within the environment, run the following in the OptLevAnalysis directory:
pip install -e .
This will install the package in developer mode, which is currently the only available method of installation.
To uninstall, run:
pip uninstall optlevanalysis
If you do not plan to make regular changes to the code and only want to use the latest stable release, run the following from the OptLevAnalysis directory:
bash update.sh
An example notebook that demonstrates how to use the FileData and AggregateData classes can be found here.
To run out-of-the-box, use the process_dataset script. The only required arguments are the path to a folder containing the raw data (including both the .h5 files and a config.yaml file) and a file prefix:
cd scripts
python process_dataset.py /path/to/data/ file_prefix
This will save the AggregateData object to the default folder with a filename generated using the path to the raw data. Some basic figures can then be produced using the make_figures script with the path to the saved AggregateData object:
python make_figures.py /path/to/aggdat.h5
opt_lev_analysis - the package from which much of this code was originally adapted.
Clarke Hardy (cahardy@stanford.edu), adapted from analysis code originally developed by Chas Blakemore and Alexander Rider.