Reproducible Spatial ML and Portfolio Projects

Record data provenance and transformations, including CRS, resampling, masking, and label decisions.

Package preprocessing and prediction operations as reusable functions and test them on a small spatial subset.

Run chunked model inference and write a standards-compliant georeferenced output raster.

Write a concise model card describing intended use, validation geography, limitations, and uncertainty.

Design a leakage-resistant ecological continuous-mapping project with explicit baselines and acceptance criteria.

Design a satellite-image classification project with spatial holdouts and class-specific acceptance criteria.

Present predictions, uncertainty, applicability, and decision-relevant limitations in a professional project report.