Skip to content

Enforce zero-gap framework comparators #49

Description

@stefan-jansen

Findings: BR-016, BR-017.

Make all selected and full cross-framework tests compare the required orders, fills, trades, PnL,
and value results. Diagnostic tolerances may describe investigation magnitude but cannot turn a
nonzero release gap into a pass.

Primary surfaces: validation comparator, scenario definitions, selected benchmark tests, retained
diff artifacts.

Acceptance:

  • The retained Zipline $9.46 difference fails the release comparator.
  • Selected cross-framework benchmarks fail when any covered result field diverges.
  • Zero-gap output passes deterministically and records exact comparison artifacts.

Dependencies: Issue 8.


Plan reference: /home/stefan/ml4t/libraries/ml4t-backtest-dev/.workspace/work/backtest-stable-release/milestone-2/plan.md - Issue 9 in milestone 0.1.0-m2 - Lifecycle and exact framework parity.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions