When trying to understand what is 'feature importance' , 'feature impact', 'feature effects', and 'prediction explanations' , .. etc. I tried to made an excel file with random number columns and a target column that is derived form previous random columns by a formula so I can manipulate the curves so as to understand them. Good idea, right? But my excel file and the curves ('feature impact', 'feature effects', 'prediction explanations', ... etc) are as boring as horizontal lines or something like that. Is there an article or example excel files existing that teaches how to play with the data and DataRobot?
You can find some datasets used in tutorials here. One commonly used is a diabetes classification project, in which the target "readmitted" is used, where we predict whether patients will be readmitted to the hospital if they are discharged. This is where you can grab the data.