Script to map predicted Bragg spot centroids to their corresponding detector panel. These panel ids can be used as additional features in learning algorithms.
To use the panel mapper,
# download the repo
git clone https://github.com/Hekstra-Lab/panel_mapper.git
# go to the repo
cd panel_mapper
# create a conda environment
conda env create -f env.yaml
# activate environment
conda activate mappanel
# install pnel mapper
pip install -e .
To confirm the installation succeeded, run the following command
map-refl-to-panel --help
This should display the different arguments.
By default, panelmapper
assumes data were collected on a Rayonix MX340-HS with binning set to 2x2.
map-refl-to-panel tests/data/subset.refl \
--x-det=3840 \
--y-det=3840 \
--x-panel=16 \
--y-panel=16 \