DataPredict

API Reference - Models - BaseModel

The base model for all machine and deep learning models.

Constructors

new()

Creates a new machine learning base model.

BaseModel.new(): BaseModelObject

Functions

setModelParametersInitializationMode()

Sets how the initial model parameters values are generated. The model parameters will be generated when train() function is run.

BaseModel:setModelParametersInitializationMode(initializationMode: string, minimumModelParametersInitializationValue: number, maximumModelParametersInitializationValue: number)

Parameters:

• initializationMode: The mode for the matrices to be initialized. Available options are:

  • Zero

  • Random

  • RandomNormal

  • RandomUniformPositive

  • RandomUniformNegative

  • RandomUniformNegativeAndPositive

  • HeUniform

  • HeNormal

  • XavierUniform

  • XavierNormal

  • LeCunUniform

  • LeCunNormal

  • None

• minimumModelParametersInitializationValue: The minimum value to be generated by the model. Only applicable to “Random” initialization mode.

• maximumModelParametersInitializationValue: The maximum value to be generated by the model. Only applicable to “Random” initialization mode.

getModelParameters()

Gets the model parameters from the base model.

BaseModel:getModelParameters(doNotDeepCopy: boolean): ModelParameters

Parameters

• doNotDeepCopy: Set whether or not to deep copy the model parameters.

Returns

• ModelParameters: A matrix/table containing model parameters fetched from the base model.

setModelParameters()

Set the model parameters to the base model.

BaseModel:setModelParameters(ModelParameters: ModelParameters, doNotDeepCopy: boolean)

Parameters

• ModelParameters: A matrix/table containing model parameters to be given to the base model.

• doNotDeepCopy: Set whether or not to deep copy the model parameters.

clearModelParameters()

Clears the model parameters contained inside base model.

BaseModel:clearModelParameters()

setNumberOfIterationsPerCostCalculation()

Set the number of iterations needed to calculate the costs. This is to save computational time.

BaseModel:setModelParameters(numberOfIterationsPerCostCalculation: number)

Parameters

• numberOfIterationsPerCostCalculation: The number of iterations for each cost calculation.

setNumberOfIterationsToCheckIfConverged()

Set the number of iterations needed to confirm convergence.

BaseModel:setNumberOfIterationsToCheckIfConverged(numberOfIterations: number)

Parameters

• numberOfIterations: The number of iterations for confirming convergence.

setTargetCost()

Set the upper bound and lower bounds of the target cost.

BaseModel:setTargetCost(upperBound: number, lowerBound: number)

Parameters

• upperBound: The upper bound of target cost.

• lowerBound: The lower bound of target cost.

setPrintOutput()

Set if the model prints output.

BaseModel:setPrintOutput(option: boolean)

Parameters:

• option: A boolean value that specifies if the output is printed.

setWaitDurations()

Set wait durations inside the models to avoid exhausting script running time.

BaseModel:setWaitDurations(iterationWaitDuration: number, dataWaitDuration: number, sequenceWaitDuration: number)

Parameters:

• iterationWaitDuration: The wait duration between the iterations.

• dataWaitDuration: The wait duration between the data calculations.

• sequenceWaitDuration: The wait duration between the token sequence.