TransmogrifAI (pronounced trăns-mŏgˈrə-fī) is an AutoML library for building modular, reusable, strongly typed machine learning workflows on Apache Spark with minimal hand-tuning
MLP was not evaluated during our experiments. In order to add it to one of our model selectors, we would need to define a parameter grid. Would you recomend a good set of parameters for it?
Otherwise, you can still use MLP directly without the model selector.
when i use MultiClassificationModelSelector i want use OpGBTClassifier Or other custom models, what should I do extends BinaryClassificationModelsToTry but it is sealed trait
MLP was not evaluated during our experiments. In order to add it to one of our model selectors, we would need to define a parameter grid. Would you recomend a good set of parameters for it?
Otherwise, you can still use MLP directly without the model selector.