Skip to content

Workflows on Models

Operations on Models

Train a parametric model

1
2
3
4
trainParametricModel[
  G, T, Q, R,
  S, M <: ParameterizedLearner[G, T, Q, R, S]](
  regParameter: Double, step: Double, maxIt: Int, mini: Double)
  • Type: DataPipe[M, M]
  • Result: Takes as input a parametric model i.e. a subclass of ParameterizedLearner[G, T, Q, R, S], trains it and outputs the trained model.

Tune a model using global optimization

1
2
3
modelTuning[M <: GloballyOptWithGrad](
  startingState: Map[String, Double], globalOpt: String,
  grid: Int, step: Double)
  • Type: DataPipe[(S, S), (D, D)]
  • Result: Takes as input a parametric model i.e. a subclass of GloballyOptimizableWithGrad, tunes it using a global optimization procedure globalOpt and outputs the tuned model.

Comments