Linear Regression about lsat HOT 2 CLOSED

Hi Nick, I was able to fix the error, I’m very grateful for your support. The error was, as you said, the LSAT read DEMs, Slopes … as continue data, but the type of pixel was in integer. So, with a geoprocess I changed the pixel type to float and LSAT worked perfectly. Thanks a lot Daniel From: BGR - Engineering Geological Hazard Assessment ***@***.***> Sent: Monday, August 1, 2022 11:23 AM To: BGR-EGHA/LSAT ***@***.***> Cc: Daniel Cuervo ***@***.***>; Author ***@***.***> Subject: Re: [BGR-EGHA/LSAT] Linear Regression (Issue #28) Hello Daniel, I was able to recreate your error by calculating Logistic Regression with "AW3D30.tif" from the test dataset as a discrete data set. LSAT tries to be smart when adding data sets to the calculation and guesses the type of parameter based on the data type: int -> categorical / discrete (e.g. lithologie) float -> continuous (e.g. slope) If a raster with n different values is loaded as a discrete data set LSAT will create n arrays where one value is 1 and the others are 0. This will use a lot of memory if there are a lot of unique values leading to the Crash you described. Make sure to select the corresponding type for continuous/discrete data sets. If all it took to fix it for you was selecting continuous for that raster and your other ones work fine the wrong automatic selection is most likely the source of the crash. If not, please share the raster data set (if possible) so I can take a closer look. Kind regards Nick — Reply to this email directly, view it on GitHub<#28 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AWAMBUF2C5FJDEWFZO7DQETVW7TTZANCNFSM55HZKMKA>. You are receiving this because you authored the thread.Message ID: ***@***.******@***.***>>