####################################################
### V6 Resource File                             ###
### generated on: Tue Dec 13 14:01:29 EST 2011   ###
###                                              ###
### This file is designed to be delivered to the ###
### translator.                                  ###
####################################################
3789=Type of Basis Function:
4801=Could not get coeff data from plug-in properties
5239=Wrong RBF data file format
6672=Number of iterations should be greater than zero.
7549=Approximation is not configured, no output parameters
8689=Could not calculate recommended number of designs
12600=Calculating RBF coefficients
13344=EBF initialization cancelled
13696=Reading RBF coefficients from coefficients data file
16130=RBF initialization returned empty data struct
18260=RBF returned unknown error code: {0,number}
18493=Failed to initialize EBF approximation
20260=Could not open RBF coefficients file.
20744=Could not evaluate standalone RBF approximation - model has not been initialized
22044=RuntimeEnv is not set, RBF technique object is not properly configured
22160=Wrong number of declared design points in the RBF data file
22921=Standalone RBF approximation has already been initialized
24292=Internal RBF data structure not found, approximation cannot be used
27076=Could not obtain EBF options from plugin properties.
29035=Maximum Iterations to Fit:
31946=RBF Technique Options
32138=Elliptical
32897=Counld not read RBF coeffients file
33465=Wrong number of inputs/outputs in the RBF data file
34147=RBF Smoothing Filter cannot be greater than {0}
37161=Could not calculate min number of designs
39882=RBF initialization failed
40913=Radial
43163=Unexpected EOF while reading RBF data file
46461=Failed to restore RBF approximation from internal data
46654=Could not read RBF coefficients file
47269=RBF configuration problem, \ncoefficients data file copy was not created, \ninitialization aborted
48973=Insufficient number of sampling points for RBF initialization: {0}
49401=Non-numeric value for "{0}" in RBF approximation technique: {1}
49975=Could not store RBF configuration in plug-in
51386=Internal error in RBF
54757=Smoothing Filter:
56233=RBF initialization cancelled
59705=There is not enough memory to build an RBF model.\n{0,number} designs may be too many.
63995=Approximation is not configured, no input parameters
64427=Could not get approx property in RBF
66911=Failed to initialize RBF approximation
67386=Could not evaluate standalone RBF approximation
72987=Must have at least 2 design points in database file for RBF initialization
75638=There is not enough memory to build an RBF model.\n{0,number} inputs may be too many.
79020=Could not evaluate RBF approximation
80019=Could not fit EBF approximation.
84493=Invalid Entry
85346=Could not reset coeff data in approx properties
92070=Invalid value for SMOOTHING_FILTER in RBF approximation technique: {0}. Value must be from 0.0 to 0.1.
92314=Could not get RBF configuration from plug-in
93359=Memory allocation failure
96588=Approximation initialization cancelled.
97172=Invalid value for {0} in RBF approximation technique: {1}.
98324=Failed to upgrade RBF data into new format
98325=EBF initialization: iteration {0} of {1}
####################################################
###   Meta Model I18N string                       #
####################################################
desc=RBF approximation technique plugin
dispname=RBF Model
longdesc=Radial Basis Function approximation is a type of neural network employing a hidden layer of radial units and an output layer of linear units. RBF approximations are characterized by reasonably fast training and reasonably compact networks. They are useful in approximating a wide range of nonlinear spaces. <p> Elliptical Basis Functions are similar to Radial Basis Function but use elliptical units in place of radial units. Compared to RBF, where all inputs are handled equally, EBF networks treat each input separately using individual weights. <p> RBF networks are characterized by reasonably fast training and reasonably compact networks. EBF networks, on the other hand, require more iterations in order to learn individual input weights and are often more accurate than RBFs. <p> Initialization of the RBF approximation requires at least 2n+1 design points to be evaluated, where n is the number of inputs.  The component being approximated can be executed multiple times to collect the required data.  Alternatively, a data file can serve as the initialization source.
techniqueoptionsdesc=<b>Smoothing Filter:</b> You can use this value to relax the requirement that the RBF approximation passes through every single data point. Its primary purpose is to smooth out noisy data. The filter operates in the Euclidian space with domains normalized to [0,1]. Clustering is performed within that domain.  <br>The maximum allowed value for the smoothing filter is 0.1.Mathematically, this means you have a maximum of 10 clustered sample points for each input/ Through research, it has been determined that at larger values it does not make sense to perform an RBF; RSM is suggested for those cases. <br><b>Type of Basis Function</b><ul><li>Radial - Creates RBF networks which are characterized by reasonably fast training and reasonably compact networks. </li><li>Elliptical - Creates EBF networks which require several iterations in order to learn individual input weights and are often more accurate than RBFs at the cost of increased initialization time. </li></ul> <b>Maximum Iterations to Fit</b> Limits the maximum number of optimization iterations used to fit an EBF model for each response based on this value.
