Training

Preparing Training Data

To enable neural network models to learn patterns and make accurate predictions, a significant amount of training data is necessary. This module does not cover the collection of reference data but instead focuses on using training data that has already been gathered. This data is typically obtained through field campaigns or by digitizing training polygons using high-resolution satellite imagery. The module highlights the importance of training data and the specific information that needs to be included in it to make it useful for model training. Additionally, it provides guidance on the necessary processing steps to prepare the data for use in the model.