XPFv4.1.0 Release Notes
Release Date
Release Overview
The focus of this release is resiliency.
-
SensibleAI Forecast:
-
Monitor System private preview
- Upgraded UI to make input of flags and filters more intuitive.
- User friendly dashboard with details pertaining to optimal model per target.
-
Bulk Data Source Refresh job
- Update more than one source table at a single time.
-
Project Configuration dashboard
- View all configurations for a given Forecast project in a single view.
- Import / export capability to seed a new project with other configurations.
-
Feature Package Manager V1
- Create custom Feature Packages that can be imported into AI Forecast Projects.
- Packages are made up from Library Features and Events.
- When importing packages, can choose the Default Package Settings option to apply the generator settings from Feature Package Manager into the project.
-
Event Transformation
- Enhancement for non-daily projects by aggregate occurrences with the settings Binary, Auto, and Count when creating events.
-
Bypass Feature Selection Events
- Ability to bypass feature selection with created events.
-
-
SensibleAI Studio:
-
New artifact visualization section
- Drastic improvements to UI for interacting with routine artifacts.
- Inclusion of C# snippets section to interact with routines via business rules.
-
-
Consumption Groups
-
Removed the option Don’t Exclude Any Models from Global Model Exclusion input
- This input is intended for users to exclude any specific model, offering that option goes against the input steps.
-
-
Xperiflow Administration Tools (XAT):
-
Bulk RSI creation
- Choose a role, scope, one to many identities to create many RSI relations.
-
Moved RSI Assignments to its own Inner Option
-
Displaying first and last name for User Identities
-
-
AI Account Reconciliation (PFC)
-
General Availability release of AI Account Reconciliation integration
- Account Reconciliation Anomaly Arena routine and related modules have been included in Xperiflow.
- Compatible with PFC-SV110-XPFv4.1.0-PV910.
-
Bug Fixes
-
SensibleAI Forecast:
-
Fixed a bug that was causing Known in Advance (KIA) features to be pruned incorrectly for valid models in FOR projects that utilized forecast overlays
-
Example: A FOR project utilizes a forecast range of 1 and 3. This project has feature X, feature X has a KIA range of 2.
- On versions < 4.0.4, the forecast range 1 models would utilize the unlagged version of feature X, however, the forecast range 3 models would use the lag(3) version of feature X. This was incorrect as they should be able to take advantage of the lag(1) version.
- On 4.1, the forecast range 1 models utilize the unlagged feature X, and the forecast range 3 models use the lag(1) of feature X.
-
-
Dependencies
OneStream Platform 9.0.0+
XBR-SV210-XPFv4.1.0-PV910
Known Issues
- If you persistently receive an
Unexpected Response from Xperiflowerrorwhen making various webservice requests, first check to make sure you are not in more than 100 OneStream Security groups. If not, then the error is unrelated, and further detail should be provided in the Error Log. - Resample routine gives the error
InvalidColumnInput() takes no keyword argumentswhen the user provides columns in the input parameters that do not exist as a column in the input file. This will not be caught by other validation. - In the Forecast section of Analysis page in Utilization, the targets pane is missing each dimension column when running a hierarchical forecasting project.
Upgrade Guide
Before upgrading to XPFv4.1.0 please be sure to upgrade directly to XPFv4.0.4 first. Once the upgrade to XPFv4.0.4 is complete, the process to upgrade to XPFv4.1.0 can be started.
If coming from a version prior to XPFv4.0.0, follow this guide:
https://datasensesoftware.atlassian.net/wiki/x/EoBJSw