0 ×

02_​Partial_​Dependence_​Pre-processing

Workflow

Partial Dependence Plot Example
partial dependencepdppdp ploticeice curvesindividual conditional expectationmachine learning interpretabilitymliMLIpreprocessingpre-processingsamplingsamplelinear samplingpartial dependence/ICE plot pre-processingcomponent
Partial Dependence / Individual Conditional Expectation (ICE) PlotThis is an example for visualizing partial dependence plot and ICE curves plot in KNIME.An XGBoost model was picked, but any model and its set of Learner and Predictor nodes can be used.- Read the dataset about wines- Partition the data in train and test- Create the samples via the apposite shared component Partial Dependence Pre-processing- Score the samples using the predictor node and a trained model- Visualize the Partial Dependence/ICE Plot and customize it directly in the View (Execute > Right Click > Open View)- Apply and Close to save the custom in-view settings Data Source This KNIME workflow uses data from Paulo Cortez, University of Minho, Guimarães, Portugal, www3.dsi.uminho.pt/pcortez A. Cerdeira, F. Almeida, T. Matos and J. Reis, Viticulture Commission of the Vinho Verde Region(CVRVV), Porto, Portugal @2009 Available at archive.ics.uci.edu/ml/datasets/wine+quality sample the data red wine data top: 90% train setbottom: 10% test settrain the modelif quality > 5 : goodelse : badtop input: curves databottom input: test datacolor by ground truthDefine boundsfor samplingof "alcohol"Partial DependencePre-processing File Reader Partitioning XGBoost TreeEnsemble Learner XGBoost Predictor Rule Engine PartialDependence/ICE Plot Color Manager Edit Numeric Domain Partial Dependence / Individual Conditional Expectation (ICE) PlotThis is an example for visualizing partial dependence plot and ICE curves plot in KNIME.An XGBoost model was picked, but any model and its set of Learner and Predictor nodes can be used.- Read the dataset about wines- Partition the data in train and test- Create the samples via the apposite shared component Partial Dependence Pre-processing- Score the samples using the predictor node and a trained model- Visualize the Partial Dependence/ICE Plot and customize it directly in the View (Execute > Right Click > Open View)- Apply and Close to save the custom in-view settings Data Source This KNIME workflow uses data from Paulo Cortez, University of Minho, Guimarães, Portugal, www3.dsi.uminho.pt/pcortez A. Cerdeira, F. Almeida, T. Matos and J. Reis, Viticulture Commission of the Vinho Verde Region(CVRVV), Porto, Portugal @2009 Available at archive.ics.uci.edu/ml/datasets/wine+quality sample the data red wine data top: 90% train setbottom: 10% test settrain the modelif quality > 5 : goodelse : badtop input: curves databottom input: test datacolor by ground truthDefine boundsfor samplingof "alcohol"Partial DependencePre-processing File Reader Partitioning XGBoost TreeEnsemble Learner XGBoost Predictor Rule Engine PartialDependence/ICE Plot Color Manager Edit Numeric Domain

Download

Get this workflow from the following link: Download

Nodes

02_​Partial_​Dependence_​Pre-processing consists of the following 47 nodes(s):

Plugins

02_​Partial_​Dependence_​Pre-processing contains nodes provided by the following 7 plugin(s):