database

Version: - Created: 07 Feb 2023

Bring Your Own Model Overview  The Vantage Bring Your Own Model (BYOM) package gives data scientists and analysts the ability to operationalize predictive models in Vantage. Predictive models trained in external tools with sample data can be used to score data stored in Vantage using the BYOM Predict. Create or convert your predictive model using a supported model interchange format (PMML, MOJO, ONNX, Dataiku, and DataRobot currently available), store it in a Vantage table, and use the BYOM PMMLPredict, H2OPredict, ONNXPredict, DataikuPredict, or DataRobotPredict to score your data with the model. For installation instructions please review the readme file above. H2O Installation Options The ability to score H2O MOJO OpenSource and Driverless AI (DAI) models has been added to the BYOM product via the H2OPredict function.  While there are no extra steps necessary to use the function to score H2O MOJO OpenSource models, there is a requirement to have a license key from H2O to be able to score the Driverless AI models.  This license key must be inserted into a Teradata table and used within the query to score H2O MOJO DAI models.     Encrypt_dai_license.sh provides an option of encrypting the H2O DAI license key before inserting it to the license table in the Teradata database. Simply follow the instructions given in the encrypt_readme.txt file to generate the encrypted license key string and insert that into the Teradata table. This script can only be run on a Linux or macOS machine.   Userexamples.zip contains the examples in the user guide. More information is contained in the userexamples_readme.txt. For additional information, please refer to the Teradata Vantage - Bring Your Own Model User Guide (B700-1111-051K)

Version: 13.10.00.05 - Created: 09 Aug 2022

You can now get a free hosted instance of Vantage for testing and development purposes at https://clearscape.teradata.com/ Download Teradata Vantage Express, a free, fully-functional Teradata Vantage database, that can be up and running on your system in minutes.  Vantage Express comes as a VM image. It can be installed on your local machine or in the cloud. In order to run this VM, you'll need to install VMware Workstation Player, VMware Fusion, VMware Server, VirtualBox, or UTM on your system. For more details, see our getting started guides. For feedback, discussion, and community support, please visit the Cloud Computing forum.

Version: - Created: 01 Aug 2022

Vantage Analytics Library Product Overview The Vantage Analytics Library (VAL) consists of a Java XSP (a SQL generator), 2 table operators (to support the KMeans and Decision Tree algorithms) and statistical test lookup tables that enable Data Scientists to analyze large datasets at scale. It operates entirely inside Vantage SQL Engine (SQLE) and no additional infrastructure is required to run. The Vantage Analytics Library behaves like a database version of an API and is compatible with Vantage SQLE 16.20 and 17.0 and 17.05. It is easy to install and requires only 40MB of space in the SQLE database for the functions. The Vantage Analytics Library provides data scientists with the following capabilities, directly in the Vantage SQLE: •    Descriptive Statistics           ○  Data Quality/Values           ○  Univariate Statistics           ○  Frequency           ○  Histogram           ○  RI Analysis (column overlap)           ○  Text Field Analysis •    Data Transformation           ○  Bin Code           ○  Formula Derivation           ○  One-Hot Encoding           ○  Null Value Replacement           ○  Recode/Rescale           ○  Sample **NEW**           ○  Sigmoid/Z-Score •    Hypothesis Tests           ○  Parametric Tests           ○  Binomial Tests           ○  Contingency Tests           ○  KS Tests           ○  Rank Tests •    Model Building & Scoring           ○  Matrix (ESSCP/Covariance/Correlation)           ○  Linear Regression           ○  Logistic Regression           ○  Factor Analysis           ○  K-Means           ○  Decision Tree           ○  Association Rules/Sequence Analysis Teradata Vantage Analytics Library software can be installed on any Vantage deployment platform, including: •    On-premises, including IntelliFlex, IntelliBase, 6xxx, and 2xxx series •    VMWare •    AWS •    Azure •    Google Cloud

Version: 13.10.00.05 - Created: 18 Nov 2021

Developer Tier is a SUSE Linux Enterprise Server (SLES) operating system and the Analytics Database packaged into a virtual container that runs in a VMware vSphere ESXI virtualized environment on third-party hardware. There are two variants of the Developer Tier including a Preconfigured version and the regular version. Developer Tier (Preconfigured) is a simplified version of VantageCore VMware Developer Tier that allows you to deploy the Analytics Database quickly and easily using the OVA template. Networking is already configured to use DHCP, and you can add storage to the system. The Preconfigured version enables you to deploy only the Database and the instructions are available in a separate user manual on this download page. The full Developer Tier consists of an OVA file template and associated property files and scripts for deploying the Teradata Database and the associated Teradata ecosystem products. Once deployed by the VMware administrator, Developer Tier can be used to evaluate the VantageCore VMware and the Analytics Database and the associated ecosystem products.   Note: You must have administrative privileges on the destination VMware environment to install and configure Developer Tier virtual machines.   The Linux OVA contains Teradata Studio, Teradata version of Jupyter Lab, Apache Superset and instructions (with a script) to install Precog on the OVA after it's installed.  To get the Jupyter access token, login as user "tdc" and run "docker ps" to view the running images. Next run "docker logs <jupyter image id>" to view the Jupyter access token.  The login and password for the Linux OVA is tdc/teradata2022 For feedback, discussion, and community support, please visit the Cloud Computing forum. Getting Started The first task is to decide whether you want to deploy the preconfigured version or the regular version of Developer Tier. Then, you must download the Developer Tier template and the manual for the version you wish to deploy. If you are deploying the full version, you'll need to also download the install scripts and desired ecosystem products from the download center. You will need the following technical requirements and components to run Developer Tier.   Component Requirements VMware ESXi version ESXi 6.x, 7.x   Datastore space for template deployment:  Teradata Database: 30 GB minimum   vCenter management • VMware administrator privileges • vCenter Standard version (must support ESXi 6.x or 7.x servers) • vCenter 6.x or 7.x   Note: There is a known issue with tdvm-init on the 17.10 Developer Tier OVA image that causes the tdvm-init script to fail. To work around this issue, run the command "mkdir -p /etc/opt/teradata/telm/tmp" and then run the command "touch /etc/opt/teradata/telm/tmp/.dttmpp" After running those commands run "python tdvm-init" to configure VantageCore VMware Developer Tier.  

Version: 13.10.00.05 - Created: 12 Aug 2021

This is a VMware OVA that contains Teradata Studio, Teradata version of Jupyter Lab, Apache Superset and instructions (with a script) to install Precog on the OVA after it's installed.  To get the Jupyter access token, login as user "tdc" and run "docker ps" to view the running images. Next run "docker logs <jupyter image id>" to view the Jupyter access token.  For feedback, discussion, and community support, please visit the Cloud Computing forum.   NOTE: The login and password for the OVA is tdc/teradata2022.