Skip to content

Conversation

@codeloop
Copy link
Member

Feature: Implement series-specific model selection using meta-features

  • Enhanced AbstractData to support subsetting of data based on specified series.
  • Introduced MetaSelector class to select the best forecasting model based on pre-learned meta-rules.
  • Updated ForecastOperatorModelFactory to utilize MetaSelector for AUTO_SELECT_SERIES.
  • Modified HistoricalData and AdditionalData classes to accept a subset parameter for filtering.
  • Added new schema option for auto-selecting series in the forecast configuration.
  • Improved data loading functions to handle subsetting effectively.
  • Refactored model report generation to ensure unique filenames for outputs.
- Enhanced AbstractData to support subsetting of data based on specified series. - Introduced MetaSelector class to select the best forecasting model based on pre-learned meta-rules. - Updated ForecastOperatorModelFactory to utilize MetaSelector for AUTO_SELECT_SERIES. - Modified HistoricalData and AdditionalData classes to accept a subset parameter for filtering. - Added new schema option for auto-selecting series in the forecast configuration. - Improved data loading functions to handle subsetting effectively. - Refactored model report generation to ensure unique filenames for outputs.
@codeloop codeloop requested a review from prasankh June 20, 2025 00:56
@oracle-contributor-agreement oracle-contributor-agreement bot added the OCA Verified All contributors have signed the Oracle Contributor Agreement. label Jun 20, 2025
@codeloop codeloop enabled auto-merge July 22, 2025 12:26
@codeloop codeloop merged commit 668593b into main Aug 20, 2025
14 of 15 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

OCA Verified All contributors have signed the Oracle Contributor Agreement.

3 participants