I am using the forest based regression tool and when I trained the model and validated it, both outputs were the same resolution as my input data (5km). But when I use the model to predict in the future, it looks like the output resolution changes or something odd is happening where the model is only predicting for pixels that aren't touching? Has anyone come across this issue and how can I fix it, or at least explain it in a manuscript?
Thank you for any insights or help on this.
I attached two figures so you can see what I am talking about.
Hi Lynne,
Yes that is exactly right, thank you for the response anyway!
Hi @KManl,
Hopefully you figured out the problem? I am wondering if the resolution of the rasters used to predict was different from those that were used to train. If that is the cause, then you could use the Resample tool to adjust the resolution of the rasters used to predict.
Lynne
Les membres connectés peuvent publier, suivre les mises à jour, et plus encore. Nouveau ici ? Inscrivez-vous gratuitement.
Find useful guides, FAQs, and documents to help you navigate and make the most of Esri Community.
