NonLinearRegression (FSharp.Stats)

NonLinearRegression Module

Namespace: FSharp.Stats.Fitting
Assembly: FSharp.Stats.dll

Types and nested modules

Type/Module	Description
GaussNewton
LevenbergMarquardt
LevenbergMarquardtConstrained	This LevenbergMarquardt implementation supports the usage of box constrains.
Table
Model
SolverOptions

Functions and values

Function or value	Description
`createModel parameterNames getFunctionValue getGradientValue` Full Usage: `createModel parameterNames getFunctionValue getGradientValue` Parameters: parameterNames : `string[]` getFunctionValue : `Vector<float> -> float -> float` getGradientValue : `Vector<float> -> Vector<float> -> float -> Vector<float>` Returns: `Model`	parameterNames : `string[]` getFunctionValue : `Vector<float> -> float -> float` getGradientValue : `Vector<float> -> Vector<float> -> float -> Vector<float>` Returns: `Model`
`createSolverOption minimumDeltaValue minimumDeltaParameters maximumIterations initialParamGuess` Full Usage: `createSolverOption minimumDeltaValue minimumDeltaParameters maximumIterations initialParamGuess` Parameters: minimumDeltaValue : `float` minimumDeltaParameters : `float` maximumIterations : `int` initialParamGuess : `float[]` Returns: `SolverOptions`	minimumDeltaValue : `float` minimumDeltaParameters : `float` maximumIterations : `int` initialParamGuess : `float[]` Returns: `SolverOptions`
`getRSS model xData yData paramVector` Full Usage: `getRSS model xData yData paramVector` Parameters: model : `Model` - xData : `float[]` - yData : `float[]` - paramVector : `Vector<float>` - Returns: `float`	Returns the residual sum of squares (RSS) as a measure of discrepancy between the data and the used estimation model. model : `Model` xData : `float[]` yData : `float[]` paramVector : `Vector<float>` Returns: `float` Example
`shouldTerminate currentValueRSS newValueRSS iterationCount currentParamGuess newParamGuess solverOptions` Full Usage: `shouldTerminate currentValueRSS newValueRSS iterationCount currentParamGuess newParamGuess solverOptions` Parameters: currentValueRSS : `float` - newValueRSS : `float` - iterationCount : `int` - currentParamGuess : `Vector<float>` - newParamGuess : `Vector<float>` - solverOptions : `SolverOptions` - Returns: `bool`	Returns true if convergence criteria are met or a user defined number of iiterations has been carried out currentValueRSS : `float` newValueRSS : `float` iterationCount : `int` currentParamGuess : `Vector<float>` newParamGuess : `Vector<float>` solverOptions : `SolverOptions` Returns: `bool` Example
`solverConverged solverOptions estParams` Full Usage: `solverConverged solverOptions estParams` Parameters: solverOptions : `SolverOptions` estParams : `ResizeArray<vector>` Returns: `bool`	solverOptions : `SolverOptions` estParams : `ResizeArray<vector>` Returns: `bool`
`standardErrorOfPrediction dOF predicted actual` Full Usage: `standardErrorOfPrediction dOF predicted actual` Parameters: dOF : `float` predicted : `float[]` actual : `float[]` Returns: `float`	dOF : `float` predicted : `float[]` actual : `float[]` Returns: `float`
`updateJacobianInplace model xData paramVector jacobian` Full Usage: `updateJacobianInplace model xData paramVector jacobian` Parameters: model : `Model` xData : `float[]` paramVector : `Vector<float>` jacobian : `Matrix<float>` Returns: `Matrix<float>`	model : `Model` xData : `float[]` paramVector : `Vector<float>` jacobian : `Matrix<float>` Returns: `Matrix<float>`
`updateResidualVectorInPlace model xData yData paramVector residualVector` Full Usage: `updateResidualVectorInPlace model xData yData paramVector residualVector` Parameters: model : `Model` - xData : `float[]` - yData : `float[]` - paramVector : `Vector<float>` - residualVector : `Vector<float>` - Returns: `Vector<float>`	Returns the residual vector, each row i contains the difference between the yEst_i and the yData_i. model : `Model` xData : `float[]` yData : `float[]` paramVector : `Vector<float>` residualVector : `Vector<float>` Returns: `Vector<float>` Example