Icon

Conformal_​regression_​by_​Redfield

Conformal regression - overview

This is a simple demo workflows that shows the main capabilities and uses cases of the Conformal Prediction extension by Redfield.
The workflow contain 33 data sets prepared for regression tasks. The workflow demonstrates 2 approaches how conformal regression can be applied:
- "simple" where only one calibration table is built;
- "advanced" where multiple calibration tables are built and used with Calibration and Conformal Prediction loops;

Normalized Conformal Regression with externally defined Sigma. Conformal Regression without normalization. Normalized Conformal Regression using the variance from the Random forest asSigma. Deploying trained models and calibration tables Training models and creating calibrationtables Deploying trained models and calibration tables Training models and creating calibrationtables CalibrationTestLearn targetLoad the path to the 33 datasets stored in the projects Data folderNode 61Node 62Node 66Node 76Node 189Node 214end CVTrainingSigma = abs(error)Learn SigmaCalibration SigmaTest SigmaSerializing modelDeserializing modelSplit data intotraining and calibrationtablesCollectingmodels andcalibration tablesPair-wise iterationover models and calibration tablesAggregating P-values(median)Node 694Node 695Node 696Node 697Sigma = prediction varianceNode 700Learn targetAggregating P-values(median)Collectingmodels andcalibration tablesDeserializing modelNode 707Node 708Serializing modelPair-wise iterationover models and calibration tablesSplit data intotraining and calibrationtablesNode 712Node 713Learn targetsignificance levelNode 745add significanceadd significanceadd significanceadd significanceadd significanceRF variance normalisationAdvanced no normalisationNo nomralisationLinreg normalisationAdvancedRF variance normalisationEnd datasetsNode 794Node 809Random Forest Predictor(Regression) Random Forest Predictor(Regression) Random Forest Learner(Regression) List Files/Folders Table Row ToVariable Loop Start File Reader Normalizer Path to String ConformalPartitioning Workflowconfiguration Row Sampling String Manipulation(Variable) Loop End Random Forest Predictor(Regression) Math Formula Linear RegressionLearner RegressionPredictor RegressionPredictor Model to Cell Cell To Model Conformal CalibrationLoop Start Conformal CalibrationLoop End Conformal PredictionLoop Start ConformalPrediction Loop End Random Forest Predictor(Regression) Random Forest Predictor(Regression) Conformal Calibrator(Regression) Conformal Predictor andClassifier (Regression) ConformalRegression ConformalRegression ConformalRegression Random Forest Learner(Regression) ConformalPrediction Loop End Conformal CalibrationLoop End Cell To Model Conformal Predictor andClassifier (Regression) Random Forest Predictor(Regression) Model to Cell Conformal PredictionLoop Start Conformal CalibrationLoop Start Conformal Calibrator(Regression) Random Forest Predictor(Regression) Random Forest Learner(Regression) Parameter OptimizationLoop Start Partitioning ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column Aggregated results Loop End Color Manager Parallelcoordinates plot Scoring thepredictions Column Filter Interval stats Normalized Conformal Regression with externally defined Sigma. Conformal Regression without normalization. Normalized Conformal Regression using the variance from the Random forest asSigma. Deploying trained models and calibration tables Training models and creating calibrationtables Deploying trained models and calibration tables Training models and creating calibrationtables CalibrationTestLearn targetLoad the path to the 33 datasets stored in the projects Data folderNode 61Node 62Node 66Node 76Node 189Node 214end CVTrainingSigma = abs(error)Learn SigmaCalibration SigmaTest SigmaSerializing modelDeserializing modelSplit data intotraining and calibrationtablesCollectingmodels andcalibration tablesPair-wise iterationover models and calibration tablesAggregating P-values(median)Node 694Node 695Node 696Node 697Sigma = prediction varianceNode 700Learn targetAggregating P-values(median)Collectingmodels andcalibration tablesDeserializing modelNode 707Node 708Serializing modelPair-wise iterationover models and calibration tablesSplit data intotraining and calibrationtablesNode 712Node 713Learn targetsignificance levelNode 745add significanceadd significanceadd significanceadd significanceadd significanceRF variance normalisationAdvanced no normalisationNo nomralisationLinreg normalisationAdvancedRF variance normalisationEnd datasetsNode 794Node 809Random Forest Predictor(Regression) Random Forest Predictor(Regression) Random Forest Learner(Regression) List Files/Folders Table Row ToVariable Loop Start File Reader Normalizer Path to String ConformalPartitioning Workflowconfiguration Row Sampling String Manipulation(Variable) Loop End Random Forest Predictor(Regression) Math Formula Linear RegressionLearner RegressionPredictor RegressionPredictor Model to Cell Cell To Model Conformal CalibrationLoop Start Conformal CalibrationLoop End Conformal PredictionLoop Start ConformalPrediction Loop End Random Forest Predictor(Regression) Random Forest Predictor(Regression) Conformal Calibrator(Regression) Conformal Predictor andClassifier (Regression) ConformalRegression ConformalRegression ConformalRegression Random Forest Learner(Regression) ConformalPrediction Loop End Conformal CalibrationLoop End Cell To Model Conformal Predictor andClassifier (Regression) Random Forest Predictor(Regression) Model to Cell Conformal PredictionLoop Start Conformal CalibrationLoop Start Conformal Calibrator(Regression) Random Forest Predictor(Regression) Random Forest Learner(Regression) Parameter OptimizationLoop Start Partitioning ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column ConstantValue Column Aggregated results Loop End Color Manager Parallelcoordinates plot Scoring thepredictions Column Filter Interval stats

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: loftuw
Created at: 2021-11-30
On NodePit since: 2022-08-24
Last update: 2024-11-17
Created with KNIME version: v4.6.2
Tags: Conformal predictionRegressionConformal regressionRedfield

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!