This repository holds the deliverables that will be given to the World Bank Group (WBG). There are several components to this that are broken up into various directories. Some of this work may be incomplete.
View the notebook demo website here
- darknet: This directory holds the Python 3 interface to Darknet. It has been modified for use by Sparkgeo.
- data: This directory holds raw and processed data to be used in the notebooks, in addition to test data that can be used to verify operation.
- lib: Compiled libraries that are necessary for the Python 3 interface. Provided are shared libraries for x86-64 Linux and Arm-based OS-X.
- model: Example model to run. This model was trained on a custom merged dataset.
- notebooks: Notebooks to perform various processing on data.
- src: Stand-alone Python 3 programs.
First clone this repository to your local environment.
To run the notebooks locally you need to install the environments for R , Quarto and Python.
To build the environment from a yml file using mamba (recommended)
mamba env create -f environment.yml
Once activating your conda environment:
conda activate wb-spk-env
You should be able to run the .ipynb notebooks.
For the provided example applications, activate the virtual environment and then run one in this fashion:
python src/count_cars.py --detection-file data/processed/test_cars.gpkg --intersect-file data/processed/beitbridge_road_mask_v2.gpkg --output-json cars.geojsonPlease see the Readme in the notebooks directory for their usage.
To replicate the R environment, you can simply install.packages('renv') and open up the R Project file and run renv::restore() to restore the renv lockfile to load in all the appropriate libraries.
Restoring the renv above will allow you to run any .qmd notebooks.
Quarto is an open-source scientific and technical publishing system.
Review Quarto's Quickstart for installation instructions.
To regenerate outputs from Quarto:
quarto render hello.ipynb --to html
quarto render hello.ipynb --to docx