Icon

KNIME_​project

Cross-Validation LoopX-Aggregator is kind of pointless in my opinion. So here an example to get more meaningful metrics for each split. Of course youneed to adjust if you are interested in different metrics.Normalization and SMOTE must only be applied to Training Data.In the real-world you would actually run a cross-valdiation on each set of parameter in case you are doing a parameter optimization.The way it is now will simply give you an idea if you can build a useable model (doesn't look like that at all) and how much the modeldepends on the input data ( a lot, very high variance between models). Node 1Node 6Node 8Node 41train / test splitNode 51Node 54Node 55Node 57Node 58Node 59Node 60Node 61Node 62Aggregated Metrics Excel Reader (XLS)(deprecated) Missing ValueColumn Filter Normalizer SMOTE Partitioning X-Partitioner Random ForestLearner Random ForestPredictor Scorer Normalizer (Apply) Number To String Loop End Row Filter Column Filter GroupBy Cross-Validation LoopX-Aggregator is kind of pointless in my opinion. So here an example to get more meaningful metrics for each split. Of course youneed to adjust if you are interested in different metrics.Normalization and SMOTE must only be applied to Training Data.In the real-world you would actually run a cross-valdiation on each set of parameter in case you are doing a parameter optimization.The way it is now will simply give you an idea if you can build a useable model (doesn't look like that at all) and how much the modeldepends on the input data ( a lot, very high variance between models). Node 1Node 6Node 8Node 41train / test splitNode 51Node 54Node 55Node 57Node 58Node 59Node 60Node 61Node 62Aggregated Metrics Excel Reader (XLS)(deprecated) Missing ValueColumn Filter Normalizer SMOTE Partitioning X-Partitioner Random ForestLearner Random ForestPredictor Scorer Normalizer (Apply) Number To String Loop End Row Filter Column Filter GroupBy

Nodes

Extensions

Links