Icon

COPA 2023 - Conformal predictive systems - simple use cases

There has been no title set for this workflow's metadata.

COPA 2023 - Conformal predictive systems - simple use cases

This workflow has been prepared as part of a paper, submitted to the COPA 2023 conference, introducing conformal predictive systems (CPS) in KNIME. The purpose is to introduce a number of simple use cases for how CPS can be used. Material in this workflow has been used in the paper.

URL: Contact Tuwe Löfström https://ju.se/en/personinfo?sign=loftuw
URL: Redfield https://redfield.ai/

Regression without conformal predictive systems Regression with conformal predictive systems Regression with conformal predictive systems, using cross validation Regression with conformal predictive systems Regression with conformal predictive systems Regression with Mondrian conformal predictive systems Learn targetNode 324Node 325Predict TestTest=30%EvaluateTest=30%Predict TestNode 332Node 333Learn targetTraining=2/3Calibration=1/3Predict CalibrationEvaluateLower=[5]Upper=[95]EvaluatePredict CalibrationTraining=2/3Calibration=1/3Learn targetNode 344Node 345Predict Test10-fold CVEnd CVLower=[5]Upper=[95]On medianResidualMedianP(cpds<REGRESSION)Test=30%Node 365Predict TestLearn targetNode 368Predict CalibrationTraining=2/3Calibration=1/3Lower=[0.5, 10, 50]Upper=[99.5, 90]Test=30%Node 374Predict TestLearn targetNode 377Predict CalibrationTraining=2/3Calibration=1/3Lower=[50]Target value=0.5Target column=REGRESSIONP(cpds<0.5)PredictionPredictionMondrianPrediction [Binned]Prediction [Binned]end MondrianEvaluateTraining=2/3Calibration=1/3Predict CalibrationNode 444Learn targetPredict TestNode 447Test=30%Lower=[5, 50]Upper=[95]Binned plotEvaluate Mondriansingle plotOn median Random Forest Learner(Regression) File Reader Normalizer Random Forest Predictor(Regression) Partitioning Numeric Scorer Partitioning Random Forest Predictor(Regression) Normalizer File Reader Random Forest Learner(Regression) ConformalPartitioning Random Forest Predictor(Regression) Conformal Scorer(Regression) Predictive SystemsRegression Conformal Scorer(Regression) Random Forest Predictor(Regression) ConformalPartitioning Random Forest Learner(Regression) File Reader Normalizer Random Forest Predictor(Regression) X-Partitioner X-Aggregator Predictive SystemsRegression Colorize Sorter Math Formula Column Rename(deprecated) Line Plot(JavaScript) Partitioning Conformal PredictiveSystems Visualization Normalizer Random Forest Predictor(Regression) Random Forest Learner(Regression) File Reader Random Forest Predictor(Regression) ConformalPartitioning Predictive SystemsRegression Partitioning Normalizer Random Forest Predictor(Regression) Random Forest Learner(Regression) File Reader Random Forest Predictor(Regression) ConformalPartitioning Predictive SystemsRegression Line Plot(JavaScript) Auto-Binner Auto-Binner (Apply) Group Loop Start ReferenceRow Filter Loop End Conformal Scorer(Regression) ConformalPartitioning Random Forest Predictor(Regression) File Reader Random Forest Learner(Regression) Random Forest Predictor(Regression) Normalizer Partitioning Predictive SystemsRegression Python View(legacy) Conformal Scorer(Regression) Python View(legacy) Sorter Regression without conformal predictive systems Regression with conformal predictive systems Regression with conformal predictive systems, using cross validation Regression with conformal predictive systems Regression with conformal predictive systems Regression with Mondrian conformal predictive systems Learn targetNode 324Node 325Predict TestTest=30%EvaluateTest=30%Predict TestNode 332Node 333Learn targetTraining=2/3Calibration=1/3Predict CalibrationEvaluateLower=[5]Upper=[95]EvaluatePredict CalibrationTraining=2/3Calibration=1/3Learn targetNode 344Node 345Predict Test10-fold CVEnd CVLower=[5]Upper=[95]On medianResidualMedianP(cpds<REGRESSION)Test=30%Node 365Predict TestLearn targetNode 368Predict CalibrationTraining=2/3Calibration=1/3Lower=[0.5, 10, 50]Upper=[99.5, 90]Test=30%Node 374Predict TestLearn targetNode 377Predict CalibrationTraining=2/3Calibration=1/3Lower=[50]Target value=0.5Target column=REGRESSIONP(cpds<0.5)PredictionPredictionMondrianPrediction [Binned]Prediction [Binned]end MondrianEvaluateTraining=2/3Calibration=1/3Predict CalibrationNode 444Learn targetPredict TestNode 447Test=30%Lower=[5, 50]Upper=[95]Binned plotEvaluate Mondriansingle plotOn medianRandom Forest Learner(Regression) File Reader Normalizer Random Forest Predictor(Regression) Partitioning Numeric Scorer Partitioning Random Forest Predictor(Regression) Normalizer File Reader Random Forest Learner(Regression) ConformalPartitioning Random Forest Predictor(Regression) Conformal Scorer(Regression) Predictive SystemsRegression Conformal Scorer(Regression) Random Forest Predictor(Regression) ConformalPartitioning Random Forest Learner(Regression) File Reader Normalizer Random Forest Predictor(Regression) X-Partitioner X-Aggregator Predictive SystemsRegression Colorize Sorter Math Formula Column Rename(deprecated) Line Plot(JavaScript) Partitioning Conformal PredictiveSystems Visualization Normalizer Random Forest Predictor(Regression) Random Forest Learner(Regression) File Reader Random Forest Predictor(Regression) ConformalPartitioning Predictive SystemsRegression Partitioning Normalizer Random Forest Predictor(Regression) Random Forest Learner(Regression) File Reader Random Forest Predictor(Regression) ConformalPartitioning Predictive SystemsRegression Line Plot(JavaScript) Auto-Binner Auto-Binner (Apply) Group Loop Start ReferenceRow Filter Loop End Conformal Scorer(Regression) ConformalPartitioning Random Forest Predictor(Regression) File Reader Random Forest Learner(Regression) Random Forest Predictor(Regression) Normalizer Partitioning Predictive SystemsRegression Python View(legacy) Conformal Scorer(Regression) Python View(legacy) Sorter

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: 2023-04-06
Last update: 2024-11-17
Created with KNIME version: v5.1.0
Tags: Conformal predictionCOPAMondrianRegressionConformal predictive systemsUncertainty quantification

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

Try NodePit Runner!