Version 3.2 (R13+) Spline Toolbox™ Software :: (Spline Toolbox Release Notes)

Spline Toolbox Release Notes

Provide feedback about this page

Version 3.2 (R13+) Spline Toolbox™ Software

This table summarizes what's new in Version 3.2 (R13+):

New Features and Changes
Version Compatibility Considerations
Fixed Bugs and Known Problems
Related Documentation at Web Site

Yes
Details below
Yes--Details listed under Compatibility Considerations, below. See also Summary
No bug fixes
No

New Features and Changes	Version Compatibility Considerations	Fixed Bugs and Known Problems	Related Documentation at Web Site
Yes Details below	Yes--Details listed under Compatibility Considerations, below. See also Summary	No bug fixes	No

New features and changes introduced in this version are

ND-Valued Data and Spline Functions

All *ap* toolbox functions now accept matrix- and ND-valued data. These functions are: csape, csapi, csaps, spap2, spapi, spaps, tpaps. All toolbox fn* functions can now handle the resulting matrix- or ND-valued spline functions.

Repeated Sites

All *ap* toolbox functions now accept data with repeated sites, hence so does the GUI splinetool. Specifically,

csape, csapi
Replace data points having the same site by their average.

csaps, spaps, spap2, tpaps
Replace data points with the same site by their weighted average, with its weight the average of the corresponding weights. These functions are concerned with smoothing or least-squares approximation.

spaps
Adjusts the given tolerance in order to make up for the resulting change in the value of the error measure.

spapi
Continues to interpret data with repeated sites as a request for osculatory interpolation

`csape`, `csapi`	Replace data points having the same site by their average.
`csaps`, `spaps`, `spap2`, `tpaps`	Replace data points with the same site by their weighted average, with its weight the average of the corresponding weights. These functions are concerned with smoothing or least-squares approximation.
`spaps`	Adjusts the given tolerance in order to make up for the resulting change in the value of the error measure.
`spapi`	Continues to interpret data with repeated sites as a request for osculatory interpolation

Updated Demos

Several demos have been enlarged to include additional illustrations and examples. You can now access the Spline Toolbox demos from the Demos tab in the Help browser, or you can run them by typing playshow demoname at the command line.

Toolbox Function Summary

Version 3.2 of Spline Toolbox provides the following:

New Functions

Function
Purpose

fnchg
A new toolbox function fnchg(fn,part,value) enables you to change the target dimension or the basic interval of the form in fn.

Function	Purpose
`fnchg`	A new toolbox function `fnchg(fn,part,value)` enables you to change the target dimension or the basic interval of the form in `fn`.

Functions with New or Changed Capabilities

Function
Purpose

csape
Input of specific end conditions for the univariate case is now the same as for the multivariate case. In both cases, you now supply specific values for endpoint conditions as additional data values. See the reference page for details.
In earlier versions, you used a fourth argument, valconds, to supply specific values for endpoint conditions in the univariate case. This argument remains valid for backward compatibility, but it may be removed in a future release, and you are encouraged to discontinue its use.

fnbrk
An expanded syntax, [out1,...,outn] = fnbrk(f,part1,...,partm), returns the part(s) of the form in f specified by part1,...,partn, where n <= m. These are the parts used when the form was put together, in spmak, ppmak, rpmak, rsmak, or stmak, but also other parts derived from these.

fnval
You can now use a multidimensional array to specify the sites at which fnval is to evaluate a spline function.

spap2, spapi
Now require that, in case of data x,y with y an array, the data value corresponding to x(j) is y(:,j), not y(j,:). See spap2, spapi Input Requirements Updated for more information.

splinetool
From the new Tools menu, you can now toggle the grid and the legend. From the Help menu, you can also look up short descriptions of technical terms used.

Function	Purpose
`csape`	Input of specific end conditions for the univariate case is now the same as for the multivariate case. In both cases, you now supply specific values for endpoint conditions as additional data values. See the reference page for details. In earlier versions, you used a fourth argument, `valconds`, to supply specific values for endpoint conditions in the univariate case. This argument remains valid for backward compatibility, but it may be removed in a future release, and you are encouraged to discontinue its use.
`fnbrk`	An expanded syntax, `[out1,...,outn] = fnbrk(f,part1,...,partm),` returns the part(s) of the form in `f` specified by `part1,...,partn`, where `n <= m`. These are the parts used when the form was put together, in `spmak`, `ppmak`, `rpmak`, `rsmak`, or `stmak`, but also other parts derived from these.
`fnval`	You can now use a multidimensional array to specify the sites at which `fnval` is to evaluate a spline function.
`spap2, spapi`	Now require that, in case of data `x`,`y` with y an array, the data value corresponding to `x(j)` is `y(:,j)`, not `y(j,:)`. See spap2, spapi Input Requirements Updated for more information.
`splinetool`	From the new Tools menu, you can now toggle the grid and the legend. From the Help menu, you can also look up short descriptions of technical terms used.

Compatibility Considerations

NaNs and Infs

Prior to Version 3.2, NaNs or Infs in the given data would propagate in the normal way through the calculations, leading to NaNs and/or Infs in the numbers that make up the output.

Now, any data point containing NaN or Inf is ignored, but a warning to that effect is printed in the command window. This affects all *ap* commands, i.e., csape, csapi, csaps, spap2, spapi, spaps, and tpaps.

csape Input Simplified

In the multivariate case, you supply specific values for endpoint conditions as additional data values. Starting with Version 3.2, you should use the same scheme in the univariate case.

In earlier versions, you used a fourth argument, valconds, to supply specific values for endpoint conditions in the univariate case. This argument remains valid for backward compatibility, but it may be removed in a future release, and you are encouraged to discontinue its use.

spap2, spapi Input Requirements Updated

For the sake of uniformity with other *ap* commands, spap2(knots,order,x,y) and spapi(knots,x,y) now require that, when y is an array, y must have as many columns as there are data sites in the vector x. That is, the data value corresponding to x(j) is y(:,j), not y(j,:). This change originally appeared in Version 3.2 Release 13+.

Prior to Version 3.2, if x was a column vector, and y was an array, then spap2 and spapi would take y(j,:), rather than y(:,j), to be the data value at x(j).

Provide feedback about this page

Version 3.2.1(R14SP1) Spline Toolbox™ Software Version 3.1.1 (R13) Spline Toolbox™ Software

© 1984-2008- The MathWorks, Inc. - Site Help - Patents - Trademarks - Privacy Policy - Preventing Piracy - RSS