Icon

Advanced Conformal Classificaiton workshop by Redfield

Pre-processing Training models and creating calibration tables Deploying trained models and calibration tables Confidence level optimization Results visualization This workflow was presented at Conformal prediction workshop on May 13th, 2020.Credits: Greg Landrum, Jeanette Prinz, Artem Ryasik.If you are interested in applying conformal prediction in your project please contact Redfieldinfo@redfield.se CHEMBL1614421De-saltingCreatingfingerprintFilteringmolecule columnDefiningtargetFilteringmissingfingerprintsTrees = 300Depth=64Calibration SetBuilding calibration tableSerializing modelTraining / Validation splitApplying modelto new dataDeserializing modelScoring conformal predictionsDefining significance levelApplying calibration tableto new predictionsSplit data intotraining and calibrationtablesCollectingmodels andcalibration tablesPair-wise iterationover models and calibration tablesAggregating P-values(median)Attaching confidence levelto metrics0.05 - 0.3Attaching confidence levelto predictionsDefining significance levelClasscountNode 657Prediction distributionby typeVisualizationsettings Row Filter RDKit Salt Stripper RDKit Fingerprint Column Filter Rule Engine Row Filter Random ForestLearner Random ForestPredictor ConformalCalibrator Model to Cell Partitioning Random ForestPredictor Cell To Model Conformal Scorer ConformalClassifier Conformal Predictor Conformal CalibrationLoop Start Conformal CalibrationLoop End Conformal PredictionLoop Start ConformalPrediction Loop End ConstantValue Column Parameter OptimizationLoop Start ConstantValue Column Prediction metricsprocessing ConformalClassifier Sparse bar charts Bar Chart Metrics andpredictions Loop End (3 ports)(deprecated) Image To Table Predictions distributionby error rate Bar chartpre-processing Bar Chart Visualizingoptimization results Table Reader(deprecated) CSS Editor Pre-processing Training models and creating calibration tables Deploying trained models and calibration tables Confidence level optimization Results visualization This workflow was presented at Conformal prediction workshop on May 13th, 2020.Credits: Greg Landrum, Jeanette Prinz, Artem Ryasik.If you are interested in applying conformal prediction in your project please contact Redfieldinfo@redfield.se CHEMBL1614421De-saltingCreatingfingerprintFilteringmolecule columnDefiningtargetFilteringmissingfingerprintsTrees = 300Depth=64Calibration SetBuilding calibration tableSerializing modelTraining / Validation splitApplying modelto new dataDeserializing modelScoring conformal predictionsDefining significance levelApplying calibration tableto new predictionsSplit data intotraining and calibrationtablesCollectingmodels andcalibration tablesPair-wise iterationover models and calibration tablesAggregating P-values(median)Attaching confidence levelto metrics0.05 - 0.3Attaching confidence levelto predictionsDefining significance levelClasscountNode 657Prediction distributionby typeVisualizationsettings Row Filter RDKit Salt Stripper RDKit Fingerprint Column Filter Rule Engine Row Filter Random ForestLearner Random ForestPredictor ConformalCalibrator Model to Cell Partitioning Random ForestPredictor Cell To Model Conformal Scorer ConformalClassifier Conformal Predictor Conformal CalibrationLoop Start Conformal CalibrationLoop End Conformal PredictionLoop Start ConformalPrediction Loop End ConstantValue Column Parameter OptimizationLoop Start ConstantValue Column Prediction metricsprocessing ConformalClassifier Sparse bar charts Bar Chart Metrics andpredictions Loop End (3 ports)(deprecated) Image To Table Predictions distributionby error rate Bar chartpre-processing Bar Chart Visualizingoptimization results Table Reader(deprecated) CSS Editor

Nodes

Extensions

Links