Packages

class SmoothingB_F extends Error

The SmoothingB_F class fits a time-dependent data vector 'y' to B-Splines.

y(t(i)) = x(t(i)) + ε(t(i)) x(t) = cΦ(t)

where 'x' is the signal, 'ε' is the noise, 'c' is a coefficient vector and 'Φ(t)' is a vector of basis functions. This version just used B-Splines.

Linear Supertypes
Error, AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. SmoothingB_F
  2. Error
  3. AnyRef
  4. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. Protected

Instance Constructors

  1. new SmoothingB_F(y: VectoD, t: VectoD, ord: Int = 4, lambda: Double = -1, method: SmoothingMethod.SmoothingMethod = ROUGHNESS, technique: RegTechnique.RegTechnique = Cholesky)

    y

    the (raw) data points/vector

    t

    the data time points/vector

    ord

    the order of the basis function (defaults to 4, cubic)

    lambda

    the regularization parameter (>= 0 or -1 to use GCV)

    method

    the smoothing method

    technique

    the factorization technique

Value Members

  1. def calcCov(yy: VectorD, k: Int = 1): MatrixD

    Calculate the correlation matrix for the basis functions.

    Calculate the correlation matrix for the basis functions.

    yy

    data vector

    k

    lag parameter for auto-covariance

  2. def d1predict(tv: VectoD): VectoD

    Predict the 1st derivative values at all time points in vector 'tv'.

    Predict the 1st derivative values at all time points in vector 'tv'.

    tv

    the given vector of time points

  3. def d1predict(tt: Double): Double

    Predict the 1st derivative value at time point 'tt'.

    Predict the 1st derivative value at time point 'tt'.

    tt

    the given time point

  4. def d2predict(tv: VectoD): VectoD

    Predict the 2nd derivative values at all time points in vector 'tv'.

    Predict the 2nd derivative values at all time points in vector 'tv'.

    tv

    the given vector of time points

  5. def d2predict(tt: Double): Double

    Predict the 2nd derivative value at time point 'tt'.

    Predict the 2nd derivative value at time point 'tt'.

    tt

    the given time point

  6. def dnpredict(n: Int, tv: VectoD): VectoD

    Predict the n-th derivative values at all time points in vector 'tv'.

    Predict the n-th derivative values at all time points in vector 'tv'.

    n

    the n-th derivative to be computed

    tv

    the given vector of time points

  7. def dnpredict(n: Int)(tt: Double): Double

    Predict the n-th derivative value at time point 'tt'.

    Predict the n-th derivative value at time point 'tt'.

    n

    the n-th derivative to be computed

    tt

    the given time point

  8. final def flaw(method: String, message: String): Unit
    Definition Classes
    Error
  9. def getBasis: DB_Spline

    Get the Basis Function object

  10. def getLambda: Double
  11. def plotBasis(tt: VectoD = t): Unit

    Predict the the basis functions

    Predict the the basis functions

    tt

    the given vector of time points

  12. def predict(tv: VectoD): VectoD

    Predict the y-values at all time points in vector 'tv'.

    Predict the y-values at all time points in vector 'tv'.

    tv

    the given vector of time points

  13. def predict(tt: Double): Double

    Predict the y-value at time point 'tt'.

    Predict the y-value at time point 'tt'.

    tt

    the given time point

  14. def residual: VectoD

    Return the vector of residuals/errors.

  15. def train(): VectoD

    Train the model, i.e., determine the optimal coeifficient 'c' for the basis functions by finding optimal Lamdba to minimize gcv.