Relative Error Rates
Edammo AutoML relative error rate: Edammo error rate / Google error rate
Training Times
Training time is the time needed for a model to "learn" patterns in the data in order to make predictions
Benchmarks
Link to full white paperEdammo Extreme AutoML vs. Google AutoML —
Relative Error Rates and Training Times
Edammo
Four standard data sets from the UCI Machine Learning Repository
Accuracy and Jaccard Indicies for Each Dataset
Human Activity Dataset
CNAE9 Business Classification Dataset
QSAR Oral Toxicity Dataset
Black boxes indicate classes that are most poorly identified with a significant advantage for the Edammo Auto ML method
Parkinson's Disease Dataset
Variance
Two dimensional visualization of the Human Activity Dataset
Visualization accomplished using Edammo's proprietary algorithm