It is well known that the efficiency of Design Exploration and Predictive Modeling solution highly depends on proper technique selection. Tedious tuning of optimization algorithm internal parameters or manual search of the proper predictive modeling technique can consume a lot of time and do not lead to success.
SmartSelection is a technique in tools for Design Exploration and Predictive Modeling in pSeven that automatically chooses the most efficient solution approach for a given type of problem and data. The principle of this technique is that the user should interact with the tool only in terms of the problem description and properties that are important and clear, not in terms of low-level details of the algorithms and techniques that user is unaware of or doesn't care about.
Provide problem description
Set of hints and options in SmartSelection helps the user to describe the problem and desired solution from his point of view, not from the algorithmic point of view. It hides techniques internal complexity so that the user could concentrate on the engineering problem itself. This opens expert level mathematics for Design Exploration and Predictive Modeling even for non-math experts.
Successful Design Exploration process highly depends on many technical aspects and may implement several Design of Experiments and Design Optimization techniques with different set of internal algorithms. That’s why choosing of one static technique for Design Exploration may not lead to the desired result.
SmartSelection uses hints about the problem provided by the user for initial navigation and chooses the algorithm automatically, while its parameters are tuned adaptively during the solution process.
In the case of performance degradation of design optimization solution process interrupts and restarts with the next suitable algorithm.
SmartSelection chooses Design Exploration technique based on problem statement
Optimization with SmartSelection vs. open algorithms.
NSGA-II, Adaptive Scalarization, SmartSelection – 280 iterations each
Selection of static technique is often not enough to build an accurate predictive model. The best model type and parameters of predictive modeling technique highly depend on a particular problem and given data. In other words, crucial information from the technical point of view might be not known beforehand.
SmartSelection for Predictive Modeling grants automatic and adaptive technique choosing:
SmartSelection setup for Predictive Modeling
Different techniques support different options and model building aspects. SmartSelection automatically navigates through all these options and chooses the best techniques for your model:
|Options||Speed vs. Quality Presets||Accuracy Evaluation||Exact Fit||Linearity Required||Tensor Structure||Internal Validation||Point Weight||Smoothing||Randomized Training||Logarithm of Outputs|
- supported - supported under certain conditions
Even with default settings SmartSelection builds predictive models of better approximation quality and in shorter time than famous open algorithms, like Scikit-learn, XGBoost and GPy. Learn more >
Accuracy profiles of different predictive modeling techniques
Training time profiles of different predicitve modeling techniques