Create prediction pipeline

[emmamendelsohn]

Some of these steps can use/adapt existing functions from the training pipeline.

• Download all data for past three months, modifying the existing data download functions to download new data only.
• Transform data, including steps to scale to 0.1 degrees, calculate lagged anomalies against stored historical values.
• Augment and aggregate into ADM regions
• Run predictions using stored model object
• Return a shapefile

See REPEL2 for the logic we used to create a data processing pipeline for prediction purposes, separate from the model training pipeline: https://github.com/ecohealthalliance/repel2/blob/dev/tar_plans/_targets_prediction_livestock.R

Note that the REPEL pipeline does not provide an example of any logic around downloading only new data, because the downloads were quick enough and the update period infrequent enough (every 6 months) that it was worth just doing a full download. For OpenRVF, we have bigger data and more frequent (monthly) updating needs.