In HANA ML 2.20, APL introduces a new tab “Local Explanations” in the time series HTML report. This new tab includes a waterfall chart showing how each component of the time series model contributed to individual forecasts. Thanks to this visualizati...
We already covered SHAP-explained models for classification and regression scenarios in a previous APL blog post, and at the time we talked briefly about the main effect of a predictor and its interaction effect with the other predictors of the model...
After completing the first part of the blog you should have a hold-out dataset in HANA dedicated to test, for both the Census case, and the California Housing case. In this second part we will build an APL model and a non-APL model on the same traini...
You need to compare an APL classification or regression model to a non-APL model, and you want to make sure the evaluation process is not flawed. This blog should help you run this comparison in a fair manner; it includes code snippets you can copy a...
Marketing campaigns need to be relevant for the recipient, and worthwhile for the sender. It is not worth sending a promotional offer about a product or service if you know that your customer will buy it anyway. It is not worth calling a subscriber t...
Here is the code to prepare the tables:create view SERIES_IN as select * from APL_SAMPLES.OZONE_RATE_LA order by "Date" asc;
create table SERIES_OUT (
"Date" DAYDATE,
"OzoneRateLA" DOUBLE,
"kts_1" DOUBLE,
"kts_1_lowerlimit_95%" DOUBLE,
"kts_...
Hi Dirk,Here is an example of anonymous block running APL forecast without Declare.DO BEGIN
header = select * from #HEADER_APL;
config = select * from #CONFIG_APL;
var_desc = select * from #DESCCRIPTION_APL;
var_...
Hello,The option "Decision Tree" is a post-modeling capability to present the trained model in a visual way.SAP Predictive Analytics 3.x always uses the Ridge Regression technique regarless of the "Decision Tree" check box that is actually not a mode...
Hi Farooq,
Correct, both uplift curves give the same AUC value of 0.03.
Using a different metric, QINI AUC, we get a value of 0.07 for the S-Learner model, and a value of 0.06 (slightly smaller) for the Y star model. I am not sure though if such di...