• funcNew
  • funcDel
  • funcEval
  • funcPlot
  • methNew
  • methDel
  • dnls1
  • dnls1e
  • tcl_func.c

    Bindings for FUNC
    

    funcNew

    TCL SYNTAX:
      funcNew  "<name>" "[list]" 
    
    TCL HELP STRING:
      Return the handle to a func.
    	gauss is a gaussian; parameters are position, sigma, and normalization
    	hubble is Hubble power law; parameters are I0 and a
    	devauc is de Vaucouleurs law; parameters are Ie and Re
    	freeman is Freeman power law; parameters are I0 and r0
    Return the handle to the new func.  
    
    TCL ARGUMENTS:
                     : 
      "<name>"     : Kind of the func to return
      "[list]"       : Optional list of VECTORs
    
    

    funcDel

    TCL SYNTAX:
      funcDel  "<func>" 
    
    TCL HELP STRING:
      Delete a FUNC and its handle
    
    TCL ARGUMENTS:
                     : 
      "<func>"     : The FUNC to delete
    
    

    funcEval

    TCL SYNTAX:
      funcEval  "<func>" "<x>" 
    
    TCL HELP STRING:
      Evaluate a FUNC at x
    Reurn the evaluated value 
    
    TCL ARGUMENTS:
                     : 
      "<func>"     : The FUNC to evaluate
      "<x>"        : Where to evaluate FUNC
    
    

    funcPlot

    TCL SYNTAX:
      funcPlot  "<pg>" "<func>" "-resFunc" "<xmin>" "<xmax>" "<npt>" 
    
    TCL HELP STRING:
      Plot a FUNC from xmin to xmax
    
    TCL ARGUMENTS:
                     : 
      "<pg>"       : Where to plot
      "<func>"     : The FUNC to evaluate
      "-resFunc"     : FUNC to subract at each point
      "<xmin>"     : Starting x value
      "<xmax>"     : Ending x value
      "<npt>"      : Number of points to plot
    
    

    methNew

    TCL SYNTAX:
      methNew  "<name>" 
    
    TCL HELP STRING:
      Return the handle to a meth.
    	chi2 is for least squares minimization
    The handle.
    
    TCL ARGUMENTS:
                     : 
      "<name>"     : Kind of meth to retrn
    
    

    methDel

    TCL SYNTAX:
      methDel  "<meth>" 
    
    TCL HELP STRING:
      Delete a METH and its handle
    
    TCL ARGUMENTS:
                     : 
      "<meth>"     : The METH to delete
    
    

    dnls1

    TCL SYNTAX:
      dnls1  "<meth>" "<func>" "<x>" "<y>" "<w>" "-mask" "-iopt" "-ftol" "-xtol" "-gtol" "-maxfev" "-epsfcn" "<diag>" "-mode" "-factor" "-nprint" 
    
    TCL HELP STRING:
      The purpose of DNLS1 is to minimize the sum of the squares of M
    nonlinear functions in N variables by a modification of the
    Levenberg-Marquardt algorithm.  The user must provide a subrou-
    tine which calculates the functions.  The user has the option
    of how the Jacobian will be supplied.  The user can supply the
    full Jacobian, or the rows of the Jacobian (to avoid storing
    the full Jacobian), or let the code approximate the Jacobian by
    forward-differencing.   This code is the combination of the
    MINPACK codes (Argonne) LMDER, LMDIF, and LMSTR.
    Returns a keyed list of func, info, nfev, njev
    
    TCL ARGUMENTS:
                     : 
      "<meth>"     : how to minimize
      "<func>"     : the function to fit
      "<x>"        : vector of x values
      "<y>"        : vector of y values
      "<w>"        : vector of weights
      "-mask"        : mask vector
      "-iopt"        : only 1 works now
      "-ftol"        : relative error in sum of squares to terminate iteration; defaults to square root of the machine precisionn
      "-xtol"        : relative error in successive iterations to terminate iteration; defaults to square root of the machine precision
      "-gtol"        : tolerence on cos of angle between fvec and any column of the Jacobian to terminate iteration; defaults to zero
      "-maxfev"      : maximum number of calls to meth; defaults to 200*(n+1)
      "-epsfcn"      : step size; defaults to machine precision
      "<diag>"     : if mode=2, n entries that serve as implicit multiplicative scale factors for the parameters
      "-mode"        : 1 variables scaled internally; 2 scaling specified by diag
      "-factor"      : used to determine initial step bound
      "-nprint"      : each nprint call to meth has iflag=0
    
    

    dnls1e

    TCL SYNTAX:
      dnls1e  "<meth>" "<func>" "<x>" "<y>" "<w>" "-mask" "-nprint" "-iopt" "<tol>" 
    
    TCL HELP STRING:
      The purpose of DNLS1E is to minimize the sum of the squares of M
    nonlinear functions in N variables by a modification of the
    Levenberg-Marquardt algorithm.  This is done by using the more
    general least-squares solver DNLS1.  The user must provide a
    subroutine which calculates the functions.  The user has the
    option of how the Jacobian will be supplied.  The user can
    supply the full Jacobian, or the rows of the Jacobian (to avoid
    storing the full Jacobian), or let the code approximate the
    Jacobian by forward-differencing.  This code is the combination
    of the MINPACK codes (Argonne) LMDER1, LMDIF1, and LMSTR1.
    Returns info:  1, 2, or 3 are OK
    
    TCL ARGUMENTS:
                     : 
      "<meth>"     : how to minimize
      "<func>"     : the function to fit
      "<x>"        : vector of x values
      "<y>"        : vector of y values
      "<w>"        : vector of weights
      "-mask"        : mask vector
      "-nprint"      : number of calls with iflag=0
      "-iopt"        : only 1 works now
      "<tol>"      : relative error to define termination