Heatmap and SHAP graph troubles

[Xin_12]

Hi guys, the first graph is the heatmap after the normalization, random forest, cross validation… But I want the x- and y-axis show the parameters, not the number. (I’ve asked about denormalizer and the parameters are string type… K-ai still can’t help) any suggestions?

Second graph is before and after I set up the SHAP nodes for SHAP graph. But still didn’t help… Is there anything wrong within the setup of the nodes? Or the process structure? Big thanks

[mlauber71]

@Xin_12 your random forest does not seem to be connected. Maybe you take a look at these examples.

https://hub.knime.com/search?tag=machine+learning+interpretability,XAI&sort=maxKudos&type=workflow

Also this book and blog might help

Enhancing Feature Engineering with SHAP Values

#SHAP #values provide us with an understanding of the impact each feature has on the model’s predictions. Read how @nilotpalc computes SHAP values with #KNIME, relying on the #AutoML component, the #SHAP loops and the #Dependence #Plot to visualize results. Enjoy the data story! PS: #HELPLINE . Want to discuss your article? Need help structuring your story? Make a date with the editors of Low Code for Data Science via Calendly → Calendly - Blog Writer