
Limited PMML Support for Analytic Models
K-means
Association rules (ARULE)
Naïve Bayes
Support is implemented with the following new analytic procedures:
PMML_MODEL
EXPORT_PMML
This new feature is described in more details in the “PMML” section of the IBM SPSS In-
Database Analytics Developer's Guide.
Logistic Regression and Generalized Linear Models (GLM)
New to this release are algorithmic procedures to support GLM. These are:
GLM
PREDICT_GLM
Detailed information on these new procedures can be found in the “Generalized Linear
Models” section of the IBM SPSS In-Database Analytics Developer's Guide.
Netezza Lua API for Analytics
Beginning with Release 2.0, Netezza Analytics includes a variant of Lua, which enables
users to take advantage of the Netezza appliance's highly parallelized environment. Lua is
an extension programming language designed to support general procedural programming
with data description facilities. It also offers good support for object-oriented programming,
functional programming, and data-driven programming. Lua is intended to be used as a
powerful, lightweight scripting language for any program that needs one. For more informa-
tion on Netezza Lua, see the IBM Netezza Lua Developer's Guide.
Perl Adapter For User-Defined Analytic Procedures
Users can extend SQL with user-defined functions (UDXs) as well as user-defined analytic
processes (UDAPs). The UDAP concept allows a user to implement a freestanding, exe-
cutable data-processing program that runs “out of process” (that is, outside the database
system), and register it in a database. UDAPs provide several important features that are
not available with a UDX. For example, a UDX must be coded in C++ but UDAPs can be
written in a variety of languages, allowing users to capitalize on their existing programming
skills. Release 2.0 provides support for writing UDAPs in the Perl language. For more infor-
mation on UDAPs and the Perl Adapter, see the User-Defined Analytics Process Devel-
oper's Guide.
Changes to nzMatrix
nzMatrix has improved the matrix multiplication procedure (GEMM) and introduced new
random number generators in this release.
14
Kommentare zu diesen Handbüchern