Designing Machine Learning Pipeline Toolkit for AutoML Surrogate Modeling Optimization

The pipeline optimization problem in machine learning requires simultaneous of structures and parameter adaptation their elements. Having an elegant way to express these can help lessen the complexity management analysis performances together with different choices strategies. With issues mind, we created AutoMLPipeline (AMLP) toolkit which facilitates creation evaluation complex using simple expressions. We use AMLP find optimal signatures, datamine them, datamined features speed-up prediction. formulated a two-stage surrogate modeling outperforms other AutoML approaches 4-hour time budget less than 5 minutes computation time.