Stands for finnish word "metsä" which means "forest". So, pymetsa is a library for forest-related spatial data processing

Before start

To use this library there is a need to place the following files in the data folder in this repository:

• arbonaUT / Raster_data
• arbonaUT / Vector_data
• arbonaUT / Lidar_variables.xlsx

Documentation

The architecture og the module is multi-layer:

• Download - access to data sources using bindings above APIs
• Preprocessing - set of functions to preprocess raw spatial data (both raster and vector)
• Sample - layer for preparing the data for neural networks training (clipping, augmentation, saving objects into files, etc.)
• Model - module for machine learning model fitting

And visualization for creating plots and save them into files.

Examples

The examples folder contain all necessary launch demo scenarios for this library

Data sources

• Landsat: https://earthexplorer.usgs.gov/. Dataset: Landsat 8-9 OLI/TIRS C2 L2. Product: Landsat Collection 2 Level-2 Product Bundle

Pymetsa as a service

Current module deployed on Heroku using the following instructions:

Deploy FastAPI on Heroku using Docker Container

heroku login

Launch docker daemon and then

heroku container:login

heroku container:push web --app pymetsa-demo

heroku container:release web --app pymetsa-demo

Swagger UI available via URL: In progress

• login: demo
• password: demo

For local launch there is a need to start launch.py

Contacts

In progress