Python HPR Detection Toolkit

Toolkit to process grasslands and evaluating their biological value

About

This package gives the necessary handles to automate the detection of unknown hpr(+) grasslands in Flanders. The packages holds different modules that can process landplots and availble data (like raster maps of height) to evaluate the presence of certain structures within the grassland that are indicative for an old/valuable grassland. The package does select which grasslands need to be inspected but only processes the given landplot. However, in the examples, more elaborate analyses workflows are included.

Features

At the moment only one type of hpr grassland is included:

• Grasslands holding ditches

Installing the package

To make sure all dependancies for the package are correct, we will work in a Virtual Environment: It's highly recommended to use a virtual environment to avoid conflicts with other projects. Here we give the instruction to do so using (mini)conda.

1. Open a (anaconda) terminal

2. Create and activate new conda environment
   Check if the python version is declared correctly (see pyproject.toml) and change if needed. Make sure to include pip in the creation of the environment! You can change the name of the new environment from INBO_microreliefto something of your liking.

conda create -n INBO_microrelief python>=3.10 pip
conda activate INBO_microrelief

3. Navigate to the package directory

cd /path/to/package/directory

4. Checkout the Git repository
   Clone the remote git repository into the package directory you navigated to. This will create a local mirror of the repository. After cloning it, go into the repository

git clone git@github.com:inbo/Python_Toolkit_HPR_Detection.git
cd Python_Toolkit_HPR_Detection

5. Install the package and dependancies
   Pip stands for Package Installer for Python. It allows you to easily install, upgrade and manage Python packages. When installing this package, pip will search automatically for dependancies declared in the pyproject.toml file and install the necessary packages. Make sure to run pip with the -e flag. This makes the project editable and ensure the default config files are found correctly.

pip install -e .

6. Verify Installation (Optional)
   You can check if the package and its dependencies were installed correctly

pip list

You should see hpr_detection_toolkit listed, pointing to your local source directory.

Usage

In the folder example_notebooks some examples of the functionality of this package are found. They use large data files such as the BWK, agriculture landusage ... Make sure to download these and point the examples to the right location using absolute paths. Some notebooks also use data that is shipped with this package.

Do not overwrite the example files or upload new data or processed data files!