Data Guru™ 3.2.0 and Demand Guru 1.0.1 Release Notes · 8 Data Guru™ 3.2.0 and Demand Guru...

Data Guru™ 3.2.0and

Demand Guru 1.0.1Release Notes

Data Guru™ 3.2.0 and Demand Guru 1.0.1 Release Notes

January 2018

Copyright © 2018 LLamasoft, Inc. All Rights Reserved.

Data Guru™ 3.2.0 and Demand Guru 1.0.1 Release Notes

Contents

Preface .................................................................................2Related Documentation .............................................................................. 2Customer Support...................................................................................... 2

Data Guru Release Notes ......................................................4What’s New .............................................................................................. 4

New Actions ......................................................................................... 4New Features ....................................................................................... 6Enhancements...................................................................................... 7Performance .......................................................................................11

Fixes.......................................................................................................12Known Issues...........................................................................................15

Demand Guru Release Notes .............................................. 18What’s New .............................................................................................18

New Features ......................................................................................19Enhancements.....................................................................................19Performance .......................................................................................20

Fixes.......................................................................................................20Known Issues...........................................................................................27

Installing Data Guru and Demand Guru .............................. 28System Requirements ...............................................................................28

Minimum System Requirements.............................................................28Prerequisite Software ...........................................................................29SQL Server Permissions........................................................................29

File Locations ...........................................................................................29Application installation folders ...............................................................30User project folders..............................................................................30User Configuration Data Folder ..............................................................30Application Data Folder.........................................................................30

Installing Data Guru..................................................................................31Installation Troubleshooting .......................................................................36

Data Guru 64-bit mode cannot use 32-bit drivers .....................................36Invalid or Expired License Message on Startup.........................................36

Preface

Preface

Welcome to the Data Guru 3.2.0 and Demand Guru 1.0.1 Release Notes. Please read through this entire document to ensure you have a full understanding of the installation procedure, as well as known issues within the application.

Related DocumentationIn addition to these release notes, documentation for Data Guru and Demand Guru includes the following: Data Guru Help Demand Guru Help Data Cube Reference Guide, a Microsoft Excel (xlsx) file

Customer SupportData Guru and Demand Guru users have access to support.llamasoft.com, which acts as a gateway to a vast array of resources, data, tools, and knowledge.

2 Data Guru™ 3.2.0 and Demand Guru 1.0.1 Release Notes

https://support.llamasoft.com

Customer Support

3Data Guru™ 3.2.0 and Demand Guru 1.0.1 Release Notes

Data Guru Release Notes


The following items are noted for Data Guru 3.2.0: What’s New Fixes Known Issues

What’s NewImprovements have been made in the following areas for this release: New Actions New Features Enhancements Performance

Additional information on these items can be found in the Data Guru Help.

New ActionsThe following new actions have been added in this release:

SAP Query ImportYou can import data into Data Guru from an SAP system, using an SAP SQ01 ABAP ‘adhoc’ query. Provides flexibility for importing data, since you can define the query in

SAP, within the SAP security confines, and then run the query in Data Guru.

Can import SAP data directly into Data Guru, eliminating the need for additional external processes

Uses native SAP security

AllocationFor a specified quantity field, you can calculate the proportional weights of hierarchical data for each level within the hierarchy, and optionally roll up data to a less granular time bucket. Optionally produces aggregated/rolled up results table(s) Can be used in conjunction with the Disaggregation action to calculate

disaggregated demand from an aggregated forecast.


What’s New

DisaggregationYou can disaggregate data down to a more granular level of a hierarchy, using aggregated data and a proportional weight table as input.

For example, you can use a demand forecast at an aggregated level and determine demand at a more granular (disaggregated) level.

Run Data Guru JobYou can run a workflow, macro, or action that exists in a project other than the project in which the action is configured. Allows you to supply values for parameters defined in the target object Allows re-use or sharing of Data Guru components by multiple projects

Time Series ForecastingMake predictions about future data values based on historical values of the data. Create and use demand data as part of a workflow that populates

supply chain models Integrate forecasting activities as a Data Guru process, eliminating the

need for an external process

Note: This action is available to Data Guru users and does not require a Demand Guru license; however, only a subset of the Demand Guru forecasting algorithms are available in Data Guru.

Demand Guru WorkbenchYou can run a workbench created in Demand Guru as a Data Guru action in a macro, workflow, or the Action Pane. Automate the execution of demand models created in Demand Guru’s

Workbench Integrate the output into a Data Guru workflow

Note: Configuration and execution of this action requires a Demand Guru license.

Data Cube ImportYou can import time series data into Data Guru from LLamasoft’s Data Cube. Use as a causal variable input for Demand Guru Workbench action Integrate retrieval of causal data, eliminating the need for an external

or manual process

Note: Configuration and execution of this action requires a Demand Guru license.



New Features

DB2 (ODBC) and Hive (ODBC) database supportSupport for the following databases connections is now available, allowing you to transfer data between the databases and Data Guru while eliminating the need for external processes. Rows and columns can be filtered, and data can be pre-aggregated on the source side.

DB2 ODBC DB2 is supported as a read/write database. Data can be imported into Data Guru from DB2 ODBC and exported

from Data Guru to DB2 ODBC. Data imported into a Data Guru transformation-mode database (such

as SQL Server or PostGreSQL) can be transformed using any of the Data Guru transforms. Data can also be written back to DB2 ODBC.

Because this connection uses ODBC, you must install appropriate ODBC drivers for your DB2 instance.

Hive ODBC Hive is supported as a read-only database. Data can be imported into

Data Guru from a Hive connection; however, Data Guru does not write back to Hive.

Data imported into a Data Guru transformation-mode database (such as SQL Server or PostGreSQL) can be transformed using any of the Data Guru transforms.

Because this connection uses ODBC, you must install appropriate ODBC drivers for your Hive instance.

Geodata Connection (previously labeled Geocoding Connection) now supports additional providersGeodata connections now support the following providers for road distance calculation and geocoding, in addition to the previously-supported Bing connection: Baidu PTV MapBox Naver

These providers, which can be more cost effective than existing alternatives, also provide flexibility for those who currently use these data sources.


What’s New

Pivot table and charts for presenting table dataPivot tables and pivot charts, in addition to the data grid, can now be used to examine table data.

Pivot grid and charts display results based on the entire table contents. This display method is in contrast to the Data Table grid, in which the number of rows shown is limited by the value you specify for the Grid Maximum. Allows you to evaluate/preview data interactively before using in a

workflow or after an action or workflow has been run Eliminates the need to transfer data to an external tool for visualization

Note: For performance reasons, results display for the pivot table grid view is limited to 2000 records, and a warning is issued if the limit is reached.

LLamasoft Customer Improvement Program participation optionYou can now contribute to future improvements of Data Guru by opting to send usage information to LLamasoft. The information is sent anonymously and used to help improve future releases of the product.

To opt into this participation program, select User Preferences > General Preferences > Advanced Settings.

For a description of the information LLamasoft collects from participants, refer to What information does LLamasoft collect? in the Online Help.

Enhancements

Data Guru 3.2.0 runs 32- OR 64-bit based on OS architectureData Guru 3.2.0 executes the 32- or 64-bit architecture version based solely on the operating system architecture of your machine, regardless of the Microsoft Office components installed.

Previously, the application loaded the Data Guru version that corresponded to the bit version of Microsoft office installed on the machine, to be compatible with the possible use of a Microsoft Access database.

In many cases, users install Access as 32-bit, and the Access database is not being used in Data Guru. Because the recommendation is to run Data Guru in 64-bit mode for performance reasons, you would have to manually start the application in 64-bit mode.

Therefore, Data Guru 3.2 loads the bit version corresponding to the OS architecture, which in most cases is the preferred 64-bit. If you require 32-bit mode (for example, to use a 32-bit mode database), you must explicitly initiate the 32-bit mode Data Guru executable from the installation folder.

Data Guru and Demand Guru file paths


http://community.llamasoft.com/Help%20Files/DG/DG_Topics/DG_Help_whatinformation.htm#dg_help_whatinformation_2125701456_1250277


If you have both Data Guru and Demand Guru licenses, the default file paths used for projects, logs and configuration files is determined based on the mode in which you started the integrated application.

You can start the integrated application either in Data Guru or Demand Guru mode, based on the program executable file you run. The file paths then include the corresponding Data Guru or Demand Guru qualifiers.

Data Guru 3.2.0 Help is web-basedBeginning with this release, help content initiated using F1 or the menu bar is hosted on the web. Prior releases of Data Guru used a local help system created during product installation.

The new help system offers improved search capabilities, and top navigation for easy insight into main topics. Its responsive format allows you to view content on multiple devices such as iPhones and tablets, in addition to traditional desktop screens.

Web-based content allows LLamasoft to perform Help maintenance and provide enhancements outside of a software release, while reducing the application footprint on your local system.

SAP BEx Import actionChanges to the SAP BEx Import action improve both functionality and usability: Attributes and properties of characteristics are available for selection as

output fields. These selections are available in the Characteristics field selection list.

The key value for a characteristic is now available for selection as an output field. The key is available as a characteristic value attribute. Note that only the Description field is returned in prior versions of DG.

The PackageSize parameter has been added and can be configured to prevent timeouts for large result sets.

Lag time during field selection/de-selection has been reduced. Execution time has been reduced, most noticeably when the action

returns a large amount of data. Pre-selection of Characteristic and “Free” Characteristic fields has been

eliminated; previous releases pre-select all Characteristic and “Free” Characteristic fields for output, requiring manual deselection.

Data Filter Logic Control action includes option for “Table Empty” conditionThis new option allows you to branch based on whether or not a table contains any rows.

Freight Data Services action allows Freight Class to be used as a fieldFreight Class can now be used as either a field or a value. In prior Data Guru versions, this property was chosen from a drop-down list and was a constant for the entire input table. This change allows the value to differ by input row.


What’s New

Data Services action supports SAP HANA input connectionYou can now execute a Data Services action using a SAP HANA connection.

Delete Rows action allows preview of deleted rowsYou can now use the Preview tab to view rows targeted for deletion, in addition to the rows to remain after deletion as in previous releases.

Delimited File Import and Excel Spreadsheet Import actions retain overridden string data typeString input/output data for a field is not transformed to an intermediary data type if the field data type is overridden on the connection definition.

Note: Imported data may appear different than that from prior Data Guru releases; for example, leading zeroes are not stripped from strings containing numeric data.

Distance Calculation action results rounded two decimal places for calculation providerRounding to two decimal places ensures consistency with Supply Chain Guru application. Results from prior Data Guru versions are integer values.

Excel Export action preserves references to targeted worksheetThe Excel Export action now retains worksheet dependencies such as formulas, pivot tables, and pivot charts. In prior Data Guru versions, references to the targeted sheet were not retained.

Excel Export action does not produce Sheet1 by defaultBecause Sheet1 is no longer produced in the Excel Export output file, manual deletion of the sheet is no longer required, such as when using the output file as input to a tool that does not recognize an empty worksheet.

File import actions no longer skip records for invalid data in unmapped fieldsThe Delimited File Import and Excel import actions no longer attempt to validate unmapped field data. Prior to this release, the application attempted to validate data in unmapped fields and skip a record if invalid data was present; for example, if an unmapped field detected as numeric contained non-numeric data.

File import actions no longer import empty rowsThe Delimited File Import and Excel File import actions no longer import rows that do not contain data. This change eliminates the need to manually remove the empty rows, either before or after import.

SAP Import action now supports structures as BAPI outputsThe SAP import action can work with BAPIs and functions that use structures as output, in addition to previously-supported output tables.



Prior to this release, only BAPI tables were supported as output, and an error was generated if an output table was not present in a BAPI/Function definition.

Improvements to connections The Create New Database Connection dialog box now lists the database

options in alphabetical order for easier location of the desired database type.

Several improvements have been made to reduce the likelihood of encountering issues related to Microsoft updates that affect the usability of some Excel files in Data Guru.

Editing the underlying file name for an Excel connection no longer automatically resets the configuration options. You can replace underlying file (for example, periodically) without

having to reconfigure the configuration options. You can also manually reset the configuration options.

The Edit Excel Connection dialog box now allows you to select the objects to be considered for analysis/consideration within Data Guru. You can search objects by name, indicating whether or not to consider worksheets and/or named ranges. Simplifies user interaction because only selected objects are

displayed in Action configuration selection lists Improves performance because only selected objects are evaluated,

reducing overhead for processing unused objects Connections pane shows the number of selected objects and the

total number of objects in the workbook You can now choose whether or not to evaluate and consider named

ranges as objects in an Excel connection, which can also reduce clutter when not using named ranges.

The Geocoding Connection has been renamed to Geodata Connection to more accurately reflect its uses in the Geocoding and Distance Calculation actions.

Special characters allowed in column and table namesSpecial characters can now be included in column and table names. The Allow special characters in column names option is set at the project level.

In releases prior to Data Guru 3.2, and in Data Guru 3.2 with the option disabled: For user-entered table and column names, you cannot enter characters

other than a-z, A-Z, 0-9, and the underscore (_). For external file imports, most special characters in column names are

replaced with underscores.

In Data Guru 3.2 with the option enabled: Special characters can be entered for table and column names that will

be created by the application. Most special characters in column names resulting from Excel and

delimited file imports are not replaced with underscores.


What’s New

Characters that are not allowed in field name entry, and still are replaced by underscores in Excel import and Delimited File import, include - those used by the various databases as object qualifiers (e.g.,:

double quote(“), square brackets ([])), forward tick (`)) backslash (\)

Additionally, no leading or trailing whitespace is allowed.

The default setting for this option is as follows: Enabled for new projects Disabled for existing projects, because once an action is edited, the

Enabled option can change column names that have special characters, and result in missing fields for downstream actions that referenced the original field name

Delimited File Import parsing improvementsChanges have been made to accommodate input data of a delimited file that is not properly quoted and/or delimited. Data Guru uses improved techniques to format the output, and attempts to limit/isolate the impact on succeeding lines of the input file.

Note: Output of some improperly formatted fields or records may appear differently than in prior Data Guru releases.

Ambiguous outer joins now treated as warningsWhen an ambiguous join is detected in a table mapping, it is now treated as a warning instead of an error as in previous releases. This change allows you to save the action without correcting the join.

Ambiguous joins are allowed (but not recommended) to be configured in actions that support table mapping, such as Select, Update, and Aggregation.

An ambiguous join may not produce the results that you expect, depending on the order in which the application processes the tables (as shown in the Execution Tree available through the Table Mapping dialog box). The execution path that Data Guru follows is based on a specific configuration, and the path is consistent between executions for that configuration.

New Data Guru program icon The icon used to represent the Data Guru application has been updated.

Performance

Delimited File Import actionPerformance improvements made to the Delimited File Import action are most noticable when the Fields can span lines option is selected. Additional improvements do not require that condition.



Access Database as output connectionTo reduce the execution time needed when an Access database is the target output connection of an action, the application now creates and writes to a local temporary directory when possible.

Action configuration - tables with large number of fieldsImprovements have been made to reduce the wait time experienced when opening and editing actions that use tables with a large number of fields.

Action configuration and macro open - large and/or complex workflowsImprovements have been made to reduce the wait time experienced when adding, removing, or modifying actions and while opening a macro or workflow. These improvements are most noticeable with large and/or complex workflows.

Note: For this improvement, Data Guru caches database table definitions. If a table definition is changed outside of DG, you must manually close and reopen the macro or workflow for the changes to be recognized.

Text Parsing action - tables with many fieldsPerformance improvements made to the Text Parsing action are most noticeable when tables used by the action have many fields.

Distance Calculation action - duplicate values presentPerformance improvements made to the Distance Calculation action are most noticeable when the tables used by the action contain a large number of records with duplicate origins and destinations.

Data Services action - execution timePerformance improvements made to the Data Services action can decrease execution time.

FixesThe following corrections have been made since Data Guru Release 3.1.7.

SQL generation for specific table mapping configurations corrected A problem in which SQL was generated incorrectly for specific table mapping configurations has been corrected.

Additionally, if a project created in a prior release contains an affected configuration, a message notifies you the first time the project is opened with Data Guru 3.2.

Note: This change can impact the result set returned from the query.


Fixes

SAP Import action corrected to address failures related to data type The SAP Import action now successfully imports using BAPI/RFCs that fail in prior Data Guru releases due to the use of certain SAP data types.

Expression Editor returns correct results when argument is calculation In the Expression Editor, the underlying SQL for arguments that are calculations now properly encloses the calculations in parentheses; previous versions may have returned incorrect results due to the missing parentheses.

This fix affects the following databases and functions: SQL Server

Mod PruneLeft PruneRight

PostgreSQL PruneLeft PruneRight

Access MOD PruneLeft PruneRight DateRollupWeek Radians Square ACOS ASIN Ceiling Degrees Log10 Power

Teradata Mod PruneLeft PruneRight Square Power

SAP HANA PruneLeft PruneRight Radians Degrees



Oracle PruneLeft PruneRight Radians Degrees

Note: This change can impact the result set returned from the query.

Adjustment action correctly uses data types for Replace function The Adjustment action now correctly interprets the Replace function when the chosen data type is Boolean, Literal, or DateTime.

Now() function returns correct results for SAP HANA The Now() function now correctly returns the date/time when used with a SAP HANA database.

Excel Import correctly imports numbers in scientific notationThe Excel Spreadsheet Import action now correctly interprets numbers represented in scientific notation that are mapped to the Integer data type.

Selection list for multi-list functions no longer cleared in computed fieldsThe selection list of elements for multi-list functions (IsNull, IsEmpty, etc.) is no longer cleared as in previous releases when used in a computed field.

DateRollupWeek function now assigns Sunday to correct week for SQL Server and Access databasesThe DateRollupWeek function correctly assigns Sunday to the prior Monday’s week, instead of the following Monday’s week as in previous releases, when used with a SQL Server or Access database.

Case-sensitive field names addressedExecution failures sometimes resulting from a difference in case between physical column names and configured column names have been corrected.


Known Issues

Known IssuesThe following issues are noted for Data Guru 3.2.

Data table view limitationsFor data displayed, if the Data Table View Row Limit is set to a value which is less than the total number of rows in the target table: You are shown an incomplete set of records You are shown an incomplete set of distinct values to filter on If you apply a filter, you are filtering the incomplete record set instead

of retrieving a new set of records based on the filter

For value filters: When you click the filter button for a column, you are shown a list of

the first 1000 distinct values for the column. If more than 1000 distinct values exist, you do not see the other

possible values, which can be misleading.

Excel import when input fields include multiple conditional formatting in Excel Excel fields with multiple conditional formatting conditions may be formatted differently when imported into DG3.2 than in prior DG versions.

Excel password-protected workbook not supported by Data Guru To use with Data Guru, you must remove the password.

Single sign-on to SupplyChainGuru.com not supported by Data Guru Consequently, if you use AD FS (Active Directory Federation Services), you cannot connect from the desktop products for cloud solving.

PostgreSQL and SAP HANA database content is case-sensitive When comparing data strings in SAP HANA and PostgreSQL, the strings do not match if the letter case does not match.

R version support for Data Guru Data Guru 3.2.0 has been tested with R Version 3.3.2. Other versions may work as well, but they have not been tested.



Each Supply Chain Guru X version compatible with specific Data Guru versionIf you use Data Guru with Supply Chain Guru X, verify the compatibility of their respective releases. Refer to Optional Software in the Supply Chain Guru X release notes for more information.

Note that Supply Chain Guru X version 1.0.2 is compatible with Data Guru 3.1.12. Check subsequent Supply Chain Guru X releases for version compatibility.

Pie charts percentage sum may be slightly larger than 100% Within the new pivot table/chart feature, the percentages that appear in pie charts may sum to slightly more than 100% due to rounding issues.

Contains, StartsWith, and EndsWith Expression Editor function should use Literal parameter data typeThe String parameter type does not work correctly with these Expression Editor string search functions. Instead, use the Literal parameter type.

The underlying SQL for these string functions does not expect the argument to be quoted. Data Guru automatically adds quotes to a String data type, and it does not work correctly. On the other hand, Data Guru does not add quotes to a Literal parameter, and so it will work correctly under these conditions.

SAP Export not functioning with SQLiteThe SAP Export action is not currently working when SQLite is selected as the input connection.

Cannot drag .xlsm files to connectionsFor .xlsm Excel connections, you cannot create a connection by dragging the file to the connections panel; you must create it from one of the Create Connection options within the application.

SQL Import action not reported when used as input connectionIn the Macro/Workflow SUMMARY pane, the SQL Import action is not included when used as an input connection.

Connection information for SQL Operation not reportedIn the Macro/Workflow SUMMARY pane, connection information for the SQL Operation is not included.

Table usage flags inconsistently updated from connection changeIn the Macro/Workflow SUMMARY pane, table usage flags are not consistently updated following a connection change.

Data Reference Check action missing output info


https://support.llamasoft.com/Downloads/SupplyChainGuru.aspx

https://support.llamasoft.com/Downloads/SupplyChainGuru.aspx

Known Issues

In the Data Reference Check action, the Info pop-up displayed using the actin’s radial menu does not properly include output information.

SAP BEx Import action selection error on scroll barIn the SAP BEx Import action, double-clicking anywhere on the scroll bar selects the highlighted line.

SAP BEx Import action SQL Preview requires refreshIn the SAP BEx Import action, you must manually refresh the output preview from the Data tab prior to previewing the SQL.

Distance Calculation action execution errors with lookup table fields In the Distance Calculation action, using lookup table fields in the filter can result in an execution error.

SQL Import action using Oracle omits warning when truncating decimal valuesIn a SQL Import action that uses an Oracle database, when a decimal value is truncated, the execution status incorrectly reports Complete instead of Complete with Warning.

Text Parsing action execution fails when SAP HANA table has no key/auto-id fieldThe Text Parsing action requires the Target table to have a unique key to preserve data integrity when updating rows.

You must create the table with a key or auto-increment field specified, or manually add the key or auto-increment in a SQL Operation transform.

Select action returns error for specific configurationThe Select action returns error during execution when configured in the following manner: The action is set up as a true cross-connection, in which input and

output connections are on different SQL Server instances. The output table is an existing table. Not all fields selected for output are mapped to the output table on the

Field Mapping tab.

Although this issue was not present in any prior release, it has been noted for this release.


Demand Guru Release Notes


The following items are noted for Demand Guru 1.0.1: What’s New Fixes Known Issues

What’s NewImprovements have been made in the following areas for this release: New Features Enhancements Performance

Additional information on these items can be found in the Data Guru Help.


What’s New

New Features

Save As added for demand modelsYou can now save a demand model under a different name, allowing you to use an existing demand model as a template for a new model.

Enhancements

Default time series name is now the field nameWhen adding a field to define a new time series, the default field name is the name of the time series, instead of a generic sequential name.

Deletion of Available Models entry now propagates changeWhen a demand model is deleted from the list of Available Models for a demand time series, that model is deleted from the Active Models list of all time series to which that model is applied, and related entries in the appropriate Model Summary tables for those demand time series are also deleted.

Output tabs automatically display data for last run demand modelWhen a demand model is run in the Workbench, the application automatically selects that model in the Model Summary tab, and the Causal Summary and LifeCycle Summary tabs also reflect data from that model.

Removal of unused Description columnThe Description column has been removed from the Model Summary table of the Time Series Forecasting Workbench.

Removal of Hourly option on Data Cube ImportThe Hourly option has been removed from Time Aggregation drop-down on the Data Cube Import action, due to its inconsistency with other parts of the application. Any existing hourly series configurations are automatically updated to daily aggregation.

Section labels for parameter lists changedTo promote cross-product consistency, the following labels have been changed in the Time Series Forecasting Workbench and Data Guru’s Time Series Forecasting action: Algorithm Parameters has been changed to Automatic Model

Parameters. Model Parameters has been changed to Algorithm Parameters.



Demand Guru 1.0.1 Help is web-basedBeginning with this release, help content initiated using F1 or the menu bar is hosted on the web. The initial release of Demand Guru used a local help system created during product installation.

The new help system offers improved search capabilities, and top navigation for easy insight into main topics. Its responsive format allows you to view content on multiple devices such as iPhones and tablets, in addition to traditional desktop screens.

Web-based content allows LLamasoft to perform Help maintenance and provide enhancements outside of a software release, while reducing the application footprint on your local system.

Forecast Horizon limit increasedOn the Time Series Forecasting Workbench, the Forecast Horizon limit is now 10,000).

Aggregation operation now appropriately executed based on causal definitionIn the Causal Time Series Definition panel, the Aggregation selection option is now recognized during execution; previously, the SUM operation was always executed regardless of the selected option.

Unsaved models deleted for Generate All ForecastsOn the Time Series Forecasting Workbench, selecting Generate All Forecasts on the Definition tab deletes unsaved models (other than the Automatic model) in the Active Models List, resulting behavior consistent with the Generate Time Series Only option.

Performance

Time Series Forecasting action executionImprovements have been made to reduce execution time of the Time Series Forecasting action.

FixesThe following corrections have been made since Demand Guru Release 1.0.

Time Series Forecasting action returns correct first value for customized datesA problem in which the Time Series Forecasting action failed to return the proper value of the first record in a series when customized dates were used for the time horizon has been corrected.

Ignore Weekends option enabled for yearly time bucketIn the Time Series Forecasting action, the Ignore Weekends option is now correctly enabled when the Time Bucket choice is Yearly.


Fixes

Ignore Weekends option produces correct values In the Time Series Forecasting action, the Ignore Weekends option is applied correctly when calculating forecast values.

Arima output displays correct model in Output Summary TableWhen the Arima algorithm has been chosen in the Time Series Forecasting action, the Output Summary Table correctly displays that choice in the model column.

Exponential Smoothing now correctly returns forecast values for Group-by time bucketsA problem in which the Exponential Smoothing algorithm was not returning all appropriate daily or monthly group-by values in the Forecast Table Output has been corrected.

Arima algorithm no longer showing duplicate parameters in configurationA problem in which the Configuration tab of the Time Series Forecasting action would sometimes display duplicate parameter names when Arima was chosen as the algorithm has been corrected.

Column names now aliased properly when duplicate names existA aliasing problem in the Time Series Forecasting action, in which use of the same column name in multiple causals and/or time series resulted in an error, has been corrected.

Inconsistencies among start dates within input groups now handled properlyAn exception encounted during execution of the Time Series Forecasting action, when start dates used for the input groups do not match, has been corrected.

Causal Time Series now correctly aggregated in the Time Series Forecasting actionA problem in which causal time series were not being aggregated correctly has been addressed in the latest aggregation methodology.

Time Series Forecasting Workbench can be saved with Demand table not in My WorkspaceA problem preventing the workbench from being saved when the table containing demand data was not in My Workspace has been corrected.

Batch IDs no longer missing from Output Table on Performance tabA problem in which Batch IDs were not included in the Output Table on the Performance tab of the Time Series Forecasting action has been corrected.



Algorithm selection preserved after changing Causal Elimination StrategyWhen an algorithm has been chosen on the Time Series Forecasting Workbench, and then a different Causal Elimination Strategy is selected, the algorithm choice is properly retained.

Time Series Forecasting action name change displayed correctly on Performance tabWhen the name of a Time Series Forecasting action is changed on the Connections tab, that change is now properly reflected on the Performance tab.

Time bucket selection list consistently ordered across productsThe choices in the selection list used to choose time bucket increments are now ordered consistently across both Data Guru and Demand Guru.

Generate HTML Report checkbox removed from Connections tabThis checkbox has been removed from the Time Series Forecasting action.

Demand Models table retains column expansionAn occasional behavioral issue noted when expanding a column of the Demand Models table, in which the column would immediately return to its original width, has been corrected.

Multiple pop-ups appearing during attempt to save incomplete workbench addressedImprovements have been made to address excessive pop-up messages appearing when an attempt is made to save a workbench that is not completely defined.

Time Series definitions correctly displayed on Configuration tab of Data Cube Import actionWhen a new Time Series is added on the Configuration tab of the Data Cube Import action, the currently selected Time Series remains displayed without having to click Get Definitions.

Slice Ratio limits now properly enforcedOn the Time Series Forecasting Workbench, only valid slice ratio values of 0-1 are permitted.

Causal Time Series name can be changed when used in WorkbenchA problem preventing the name change of a causal time series that is being used in a Time Series Forecasting Workbench has been corrected.


Fixes

Legend graph labels updated with demand model name changeWhen the name of a demand model is changed, the Seasonality and Trend and Show Outliers labels appearing on the graph legend are properly updated.

Data point graph colors now appear on Show Outliers legendThe graph legend now properly identifies colors used to represent different data point types.

Graph data now aligned correctlyGraph data on the Time Series Forecasting Workbench for two series displayed when Growth Fit and Curve is selected for the Life Cycle demand model type is now aligned properly. Other graph alignment problems have also been addressed.

Validation now applied when attempting deleting a workbench used in a Data Guru actionA validation problem that allowed deletion of a Time Series Forecasting Workbench used in Data Guru’s Demand Guru Workbench action has been corrected; however, this deletion can occur in two conditions described in Known Issues.

Auto-detect of time horizon now works properly for custom date rangesA problem preventing successful auto-detection of the time horizon after specifying a custom date range has been corrected.

Data Cube Import action now properly removes all tables using Delete Table after Macro Execution optionWhen the Delete Table after Macro Execution option is selected in a Data Cube Import action that is used in a macro with multiple Time Series Forectasting actions, the table is properly removed for all Time Series Forecasting actions in the macro when it is run.

Long text values in fields now use ellipses to indicate longer nameWhen text strings in a field exceed the width of a field, the string is truncated and accompanied by an ellipses, and a tooltip allows easy display of the full text without having to resize the column.

Demand Modeling tab defaults to first demand time seriesAfter selecting either the Generate Time Series Only or Generate All Forecasts option on the Definition tab of the Time Series Forecasting Workbench, the Demand Modeling tab now correctly defaults to the first demand time series.

Data Cube exception resolved for blank connection values



An exception encountered when selecting the Configuration tab if the Data Services connection values are blank has been corrected.

Correct algorithm displayed for Simple Moving AverageWhen Simple Moving Average is the selected algorithm, it is correctly identified in the Summary Table.

Time Series Forecasting Workbench lists only valid connection choicesBecause non-SQL Server databases are not supported as valid Workbench connections, the Connections tab of the Time Series Forecasting Workbench no longer lists those connections.

All Causal Time Series list scrolling issues resolvedA behaviorall correction made on the Definition tab of the Time Series Forecasting Workbench allows you to scroll properly through the All Causal Time Series list using the scroll bar.

editing of a Causal Time Series Definition resulted in breaking of or in inaccuracies in objects in which they were used --> disable renaming or removal of a causal from a Causal Time Series Definition when the definition is in use by a saved object

Editing of causal tme series definition disabled when in useThe application now correctly disables editing/renaming of a Causal Time Series Definition if it is being used by a saved object.

Clear All now working properly in Output paneA problem in the Output pane of the Time Series Forecasting Workbench, in which data removed from the pane using Clear All would return after a subsequent execution, has been corrected.

Auto Arima exception resolvedAn exception encountered when using the Auto Arima algorithm on a specific demand time series has been corrected.

Changes to demand model associated with multiple time series now saved correctlyA problem in which attempting to change or rename a demand model that was associated with more than one demand time series could result in a database error has been corrected.

Save prompt when closing application now working correctlyWhen you attempt to close the Demand Guru application using the application window’s Close icon X, and changes have been made but not saved, you are prompted to save your changes.

Graph display icon working correctly


Fixes

A problem in which deselecting the graph display icon for an Active Causal Time Series in the Demand Model panel could unintentionally clear the contents of the table associated with the graph has been corrected.

Connection issues for Time Series Forecasting Workbench addressed Changing the input connection now properly clears the selected table. Changing the input table after configuring a time series definition no

longer causes an exception.

Generate All Forecasts now correctly updates time series forecasts based on new time bucketWhen forecasts have been generated and the time bucket is then changed (for example, from monthly to weekly), subsequent forecasts properly reflect the new bucket increment.

Diagnostic files no longer written by defaultDiagnostic files created during demand model execution are no longer written by default; instead, diagnostic files must be enabled by LLamasoft support.

Special characters in names now properly handledThe use of special characters in demand models or causal time series names no longer results in an exception when running models or executing actions.

Demand models from multiple workbenches cannot be executed concurrentlyA problem that allowed execution of demand models from more that one workbench simultaneously has been corrected; you can only execute one workbench at a time.

Data Services connection correctly created from Data Cube Import actionIf a Data Services connection exists, and another one is then created from the Connections tab of a Data Cube Import action, the new connection become the default selection for the open import action, and subsequent selection of the older connection in that open action does not improperly remove the new one from the selection list.

Corrections for causal time series behaviorSeveral changes have been made to address problems encountered in configuration and execution of causal time series definitions.

Corrections to demand modeling algorithmsSeveral changes have been made to address problems encountered when using algorithm parameters.



Duplicate demand model name issues addressedSeveral changes have been made to address problems that allowed duplicate names for demand models; unique model names are now properly enforced.

Clear table(s) before execution choice removed from Time Series Forecasting WorkbenchIn the Output options for the Time Series Forecasting Workbench, the Clear table(s) before execution option has been removed, because the workbench output does not have an existing table to clear.


Known Issues

Known IssuesThe following issues are noted for Demand Guru 1.0.1.

Assigned Causal definitions not shown in Causal Time Series to consider following Demand Guru upgradeWhen a project is upgraded from Demand Guru version 1.0, the assigned causal definitions in the Time Series Forecasting Workbench do not appear properly in the Modeling tab’s Causal Time Series to consider list; instead, they are included in the Active list and must be moved manually.

Time Series Forecasting Workbench exception when database is emptyIn the Time Series Forecasting Workbench, an exception is encountered when opening the workbench if the database is empty with no tables.

Time Series Forecasting Workbench applied to Demand Guru Workbench can be deletedUnder certain conditions, it is possible to delete a Time Series Forecasting Workbench that has been linked to a Demand Guru Workbench action in Data Guru: The workbench is applied to an action that is saved to the Action pane. The workbench is applied to an action that exists in a workflow or

macro, but the workflow or macro is not open.

Algorithms generate exception for Simple Moving Average algorithm parameter

In the Time Series Forecasting Action and Data Guru’s Demand Guru Workbench action, the Simple Moving Average algorithm generates an exception if the number of data points is less than the value of the Moving Average Type parameter.

Time Series Forecasting Workbench demand model rename does not commit change for EnterIn the Time Series Forecasting Workbench, using the Enter key to save the renaming of a demand model has no effect on the model name.

Time Series Forecasting Workbench deletion shows no confirmationWhen a Time Series Forecasting Workbench is deleted, no confirmation is displayed.

R version support for Demand Guru Demand Guru 1.0.1 requires R Version 3.3.2. It will not work with other versions of R.


Installing Data Guru and Demand Guru

Comp ce

Purpo

Data r


This chapter provides all the information needed for successful installation of Data Guru and Demand Guru, which are both 64-bit applications: System Requirements File Locations Installing Data Guru Installation Troubleshooting

System Requirements

Minimum System RequirementsData Guru and Demand Guru can run on most modern desktop and laptop machines with few system limitations. Additional system performance allows for faster processing of larger data sets. These applications can be run on a machine with the following minimum specifications: 2.6 GHz Intel i5 4 GB RAM 100 GB disk space free (1 GB for installation) 1024X768 screen resolution Supported Microsoft Windows Environments

Windows 7 (64-bit) Windows 8 (64-bit) Windows 10 (64-bit)

Internet connectivity required for certain functionality

While these are the absolute minimum requirements to install and run Data Guru and Demand Guru, it is strongly recommended that you use more advanced systems.

Data Guru allows users to transform their data on a cloud or local server. If the data transformation process is going to happen on the local client machine, hardware specs should be increased. A solid state hard drive and increased memory are strongly recommended.

Recommended System Specifications

onent Basic Client Standard Client

Low End Server

Standard Server

High PerformanServer

se User client machine for day-to-day work

Dedicated data server to be accessed by client machine

(GB) * 4 or Less 6-10 10-30 30-60 75 or greate


File Locations

CPU

Memo ter

Hard ID

DBMS

OS

Comp ce

* The relative size of the database on the client machine which is supporting the data transformation process.

Prerequisite Software R-Install3.3.2 Microsoft Visual C++ 2013 Redistributable Package Microsoft .NET Framework 4.6.1 Microsoft SQL Server 2014 Express RTM LocalDB (x64)

You can use a variety of database systems. This install provides a copy of SQL Server LocalDB that can be used with the applications.

Microsoft .NET 3.5 Microsoft .NET Framework 3.5 SP1 (Windows Feature) Microsoft .NET Framework 3.5 SP1This is required by the SQL Console app distributed by LLamasoft.

SQL Server PermissionsIf your workspace database is on a SQL Server instance that is not a SQL Server LocalDB or SQL Server Express instance, then the following permissions are needed for the workspace database: db_ddladmin db_owner

File LocationsSeveral file locations are used by the Data Guru and Demand Guru applications. These are created for each application during installation and include the following: Application installation folders Application data Project folders Configuration file

Intel I7 Intel I7 Single Xeon processor

Single Xeon processor

Dual Xeon Processor

ry (GB) 6 8 32 64 128 or grea

Drive 500 GB HDD 240 GB SSD RAID Array RAID Array SAN/SSD RAArray

SQL Express SQL Standard SQL Standard SQL Enterprise

Win 7 or 8 (64-Bit)

Windows Server 2008 / 2012

onent Basic Client Standard Client

Low End Server

Standard Server

High PerformanServer



Each product (Data Guru and Demand Guru) has its own copy of each directory path. The path that gets is used is determined based on which application you open first in a single worksession. Once you open the application for a product, that same folder is used for the duration of the session, even if you toggle between applications.

Application installation foldersDuring installation, Data Guru and Demand Guru application installation files are placed under the appropriate Program Files folder on the system drive: C:\Program Files\LLamasoft\DataGuru C:\Program Files\LLamasoft\DemandGuru

User project foldersDuring initial start-up, a set of user project folders are created, which by default are placed in one of the following locations depending on the application being opened: C:\Users\current.user\Documents\LLamasoft\DataGuru\Projects C:\Users\current.user\Documents\LLamasoft\DemandGuru\Projects

These locations are always offered when creating a new project; however, you can override the default and select another location.

User Configuration Data FolderApplication configuration information for each application is stored in the user.config file located in the application’s AppData\Local folder of the current user: C:\Users\<current.user>\AppData\Local\LLamasoft\DataGuru C:\Users\<current.user>\AppData\Local\LLamasoft\DemandGuru

Typical user configuration information includes the most-recently-used lists for projects and last accessed folder locations. This information is independent of projects.

Note: The AppData folder is a hidden folder. You must disable the Hidden attribute on this folder to view its contents.

Application Data FolderData Guru and Demand Guru each use a folder common to all users of their respective application: C:\ProgramData\LLamasoft\DataGuru\Diagnostics C:\ProgramData\LLamasoft\DemandGuru\Diagnostics

Currently, this folder is used solely to store the Diagnostics sub-folder for the application’s diagnostic output, which is written to a file named ErrorLog.xml. The application writes to this file when it encounters an unexpected condition or when an internal program exception occurs. Currently, you must send this file manually to LLamasoft if you require


Installing Data Guru

assistance in resolving a program error. Later application versions may automatically send this file to LLamasoft as errors occur. The file is formatted in XML and can be programmatically parsed.

Note: This file contains only error information; trace information is not included. Whenever this file has contents, it is of interest to the development team for quality purposes.

In the future, additional data common to all users of the application will reside in the Application Data folder.

Installing Data Guru1. Download the Data Guru installation file DGuru.exe from

support.llamasoft.com.2. Run Data Guru and Demand Guru 64.exe as an administrator.3. If prompted to allow the installer to execute, click Yes.

The installer first checks to see whether any required items must first be installed before beginning the Data Guru and Demand Guru installation. If any missing items are detected, they are listed as in the following example.

4. Click Install to install the required items. Note that you may be prompted by a User Account Control screen to allow the items to be installed. After providing confirmation, it can take several minutes as the items are installed, and a series of screens display indicating the status of each item during this process.



During this part of the installation, you can optionally install Microsoft SQL Server 2014 Express RTM LocalDB (x64). We recomment that you select Yes when prompted if you do not already have a database installed.Once all the prerequisite items are installed, the InstallShield Welcome screen for Data Guru and Demand Guru displays.

If one of the prerequisite items being installed is Microsoft .NET Framework 4.6.1, you may receive a message that the installation of .NET Framework 4.6.1 failed, with a prompt to continue. We recommend that you select Yes to continue with the installation; after the

other components are installed, you are prompted to reboot. If you select No, you should manually reboot at this point (even though there

is no prompt to do so), and restart the installer as an Administrator. If you do not reboot before restarting the installer as an administrator, the .NET installation is not recognized, and you must start the installation from the beginning.



5. Click Next to display the License Agreement, and indicate your acceptance of the license terms.

6. Click Next.7. When prompted to choose a destination location for the installed files,

accept the default location, or click Change to choose an alternative location. However, you should only install Data Guru on your C: drive.



8. Click Next to display the Ready to Install... screen, and then click Install.

9. At this point, if a User Account Control screen again prompts you for permision to make changes to your computer, click Yes.Otherwise, the install begins and, during the installation, a progress bar indicates the status.An additional screen displays as the R packages are installed into the Program Data folder.



10.When the installation has completed, indicate whether you want to see the Windows Installer log, and click Finish to exit the InstallShield Wizard.

The application is now installed as shown:



Installation Troubleshooting

Data Guru 64-bit mode cannot use 32-bit driversThis includes 32-bit Microsoft Office and 32-bit Access drivers that are installed after 64-bit Microsoft Office is installed.Data Guru runs as a x64 application and cannot use the x86 (32-bit) version of the Access database driver. Because Data Guru is not able to use the Access database, it displays a warning that the x64 Access driver cannot be found.

A similar warning is issued by Data Guru for 32-bit ODBC drivers.

Invalid or Expired License Message on StartupWhen starting Data Guru, the following message is displayed if a valid license is not found or if an expired license is found:


Installation Troubleshooting

Should this occur, click OK to open the License Manager.

On the License Manager screen, click the Register button and supply a valid license file.


Data Guru™ 3.2.0 and Demand Guru 1.0.1 Release Notes · 8 Data Guru™ 3.2.0 and Demand Guru...

Documents

Transcript of Data Guru™ 3.2.0 and Demand Guru 1.0.1 Release Notes · 8 Data Guru™ 3.2.0 and Demand Guru...