ML challenge to predict soil parameters from hyperspectral images.
-
Updated
Oct 17, 2022 - Jupyter Notebook
ML challenge to predict soil parameters from hyperspectral images.
Similar to LuckyHTML, but for Hyperview (https://hyperview.org/).
Project starter for hyperview, an hypermedia based mobile app framework.
My mobile apps
Soil P, K, Mg & pH from 150-band hyperspectral images (ESA HYPERVIEW challenge): mask-aware preprocessing, spectral derivatives, per-target feature selection, RF/XGBoost/kNN with leak-free repeated CV, plus a Streamlit app.
Reference server-driven TODO app built with Django, dj-hyperview, Hyperview, React Native, and Expo.
To associate your repository with the hyperview topic, visit your repo's landing page and select "manage topics."