Optimization Toolbox Release Notes |
Upgrading from an Earlier Release
This section describes the upgrade issues involved in moving from the Optimization Toolbox 2.1.1 to Version 2.2.
If you are upgrading from a version earlier than 2.1.1, then you should also see Upgrading from an Earlier Release in the Optimization Toolbox 2.1. Release Notes.
fsolve Now Defaults to New Medium-Scale Method
In keeping with the new default trust-region dogleg algorithm, fsolve
now defaults to the medium-scale method. The default for the 'LargeScale'
parameter is changed to 'off'
.
Using the Large-Scale Method
If you want to continue using the large-scale fsolve
method, which was the default in Versions 2.0 and 2.1, you must explicitly specify the large-scale method using code similar to
Using the Gauss-Newton Method
In prior releases to get the Gauss-Newton method you did:
Note
Strictly speaking, it is unnecessary to set 'LargeScale' to 'off' , unless you have previously set it 'on' .
|
Using the Levenberg-Marquardt Method
In prior releases to get the Levenberg-Marquardt method you did:
Use playshow Command to Run Slideshow Style Demos
Starting in Release 13, to run slideshow style demos from the command line, you must use the playshow command. For example,
You can continue to run other styles of demos from the command line by typing just the demo name. Optimization Toolbox demos that are affected: circustent
, molecule
, optdeblur
.
Major Bug Fixes | Optimization Toolbox 2.1.1 Release Notes |