'How to start a process of making MAE maps of my random forest model using my covariates and a predictor (clay)

I am working in R and I have my random forest model done and "cv".

What I am doing is predicting soil texture of sand, silt, and clay using 10 covariates.

Is there a way to create a MAE map of e.g., clay?

I think I have to raster brick my covariates then I do not know what to do from there.

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Solution	Source