XPFv4.3.0 Release Notes
Release Date
Release Overview
The theme of this release is simplifying end-user adoption, reducing use case maintenance complexity, and decreasing the complexity to create new use cases.
Forecast Agent (Beta)
- Ask questions about SensibleAI Forecast projects in plain English. Get grounded answers about forecast accuracy, key drivers, model explanations, and job status - no dashboard navigation required.
- Access the Forecast Agent
Monitor System
- Monitor System - Rule Based Model Selection & Alerting
Feature Forecasting
- Optional Feature Forecasting capability enables future feature values to be automatically estimated when source data does not extend through the forecast horizon, helping preserve predictive signals for downstream forecasts within SensibleAI Forecast. Additional details are available in the Feature Forecasting documentation.
Webservice overhaul
- Added FastAPI to handle async requests for agentic capabilities, FastAPI now handles all resources besides Dash webapps and related auth endpoints (Flask).
- Fully reverse proxy config with IIS
React Webapp Deployment
- New service / infrastructure to support the Forecast Agent
Feature Selection Deduplication
- Previously we had ran feature selection for each individual xperiment kernel independently, this resulted in lots of duplication of work because any models with the same forecast range and input features should select the same set of features (although this isn’t fully deterministic so it’s possible to get different results)
- Now feature selection is only ran once for each forecast range + input feature set and results are copied to all kernels/datafilters that are included in each unique combination, this saves significant time on pipeline runs.
DataForge Routine
- Internal tool for demo engineers and engineers to create robust testing time series dataset
Bug Fixes
Explanation of Reconciliation Strategies in Hierarchical Forecasting
- Full descriptions of each reconciliation strategy right at the implementation source instead of searching through documentation.
Defaulted Confidence Intervals
- Confidence intervals will now be enabled at all times.
Dynamic Scaling of Summation Numbers
- Easier readability of large numbers throughout the solution.
Simulator in Utilization
- Users can now utilize simulator actions in utilization and not just prior to the pipeline.
Dependencies
OneStream Platform 9.3.0+
AIS Dependency Bundle SV112
Known Issues
-
Forecast Agent - inability to download or copy a table when fully expanded.
- Workaround: do not expand the table to full screen - the copy and download capability will work as expected.
-
Data refresh with multiple datasources may produce database integrity errors
-
When a scheduled or manual data refresh runs against a project with multiple datasources, the system processes each datasource's downstream data update independently and in parallel. The concurrent tasks can attempt to insert the same records simultaneously, resulting in
IntegrityErrorfailures. Affected experiment builds may have incomplete or missing transformed data after the refresh completes. -
Workarounds:
- Option A: Run datasource refreshes one at a time (single-datasource update) rather than using the bulk multi-datasource refresh.
- Option B: Create a support ticket to have the AI Support team add a task handler to prevent this issue from ever occurring.
-
Upgrade Guide
(XPF) Xperiflow 4.3 Upgrade Guide