Learns a Gaussian process regression model.
| Parameter | Description |
|---|---|
| applyRowOrder | Specifies that you wish that the action uses a prespecified row ordering. This requires using the orderby and groupby parameters on a preliminary table.partition action call. Alias: reproducibleRowOrder. Default: FALSE. |
| attributes | Changes the attributes of variables used in this action. Currently, attributes specified on the inputs and nominals parameter are ignored. Alias: attribute. Sub-parameters: format, formattedLength, label, name (required), nfd, nfl. |
| autoRelevanceDetermination | When set to True, use automatic relevance determination in the kernel function. Alias: ard. Default: FALSE. |
| display | Specifies a list of results tables to send to the client for display. Sub-parameters: caseSensitive, exclude, excludeAll, keyIsPath, names, pathType, traceNames. |
| fixInducingPoints | When set to True, fixes inducing points in the optimization. Default: FALSE. |
| fixKernelParmFirstIter | When set to True, fixes kernel parameters in the first iteration. Default: FALSE. |
| inputs | Specifies variables to use for analysis. Alias: input. Sub-parameters: format, formattedLength, label, name (required), nfd, nfl. |
| jitterMaxIters | Specifies the maximum number of iterations for jitter Cholesky decomposition. Default: 10. Minimum value: 0. |
| kernel | Specifies the kernel function type for Gaussian distributions in the Gaussian process regression model. Default: RBF. Options: LINEAR, MATERN32, MATERN52, PERIODIC, RBF. |
| nInducingPoints | Specifies the number of inducing points. Default: 100. Minimum value: 2. |
| nloOpts | Specifies the optimization options. Alias: optimizer. Sub-parameters: algorithm, optmlOpt, printOpt, sgdOpt, validate. |
| outInducingPoints | Specifies the output data table in which to save the estimated mean and standard deviation at inducing points. Sub-parameters: caslib, compress, indexVars, label, lifetime, maxMemSize, memoryFormat, name, promote, replace, replication, tableRedistUpPolicy, threadBlockSize, timeStamp, where. |
| output | Specifies the output data table in which to save the scored observations. Sub-parameters: casOut (required), copyVars, role. |
| outputTables | Lists the names of results tables to save as CAS tables on the server. Sub-parameters: groupByVarsRaw, includeAll, names, repeated, replace. |
| outVariationalCov | Specifies the output data table in which to save the estimated variational distribution's covariance matrix at inducing points. Sub-parameters: caslib, compress, indexVars, label, lifetime, maxMemSize, memoryFormat, name, promote, replace, replication, tableRedistUpPolicy, threadBlockSize, timeStamp, where. |
| partByFrac | Randomly assigns specified proportions of the observations in the input table to training and validation roles. Observations are logically partitioned into disjoint subsets for model training, validation, and testing. Sub-parameters: seed, test, validate. |
| partByVar | Specifies the variable in the input data whose values are used to assign roles to each observation. Observations are logically partitioned into disjoint subsets for model training, validation, and testing. Sub-parameters: name (required), test, train, validate. |
| saveState | Specifies the output data table in which to save the state of the Gaussian process regression for future scoring. Sub-parameters: caslib, label, lifetime, memoryFormat, name, promote, replace, tableRedistUpPolicy. |
| seed | Specifies the seed value for random number generation in initializing parameters and clustering. Default: 0. |
| table | Specifies the settings for an input table. Sub-parameters: caslib, computedOnDemand, computedVars, computedVarsProgram, dataSourceOptions, importOptions, name (required), singlePass, vars, where, whereTable. |
| target | Specifies the target variable to use for analysis. |
| useSimpleInit | When set to True, uses simple parameter initialization for the optimization. Default: TRUE. |