Skip to content

Added ChemCensor evaluation subpackage - #272

Draft
Darkxzie wants to merge 1 commit into
devfrom
chemcensor
Draft

Added ChemCensor evaluation subpackage#272
Darkxzie wants to merge 1 commit into
devfrom
chemcensor

Conversation

@Darkxzie

Copy link
Copy Markdown
Collaborator

Added deepretro/evaluation: an offline, research-facing harness that scores ranked single-step precursor predictions with the official ChemCensor engine and aggregates PT-Max / PT-Top-K plausibility metrics.

Description

Fix #(issue)

Type of change

Please check the option that is related to your PR.

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds functionality)
  • Breaking change (fix or feature that would cause existing functionality to not work as expected)
    • In this case, we recommend to discuss your modification on GitHub issues before creating the PR
  • Documentations (modification for documents)

Checklist

  • I have performed a self-review of my own code
  • I have commented my code, particularly in hard-to-understand areas
  • I have made corresponding changes to the documentation
  • I have added tests that prove my fix is effective or that my feature works
  • New unit tests pass locally with my changes
  • I have checked my code and corrected any misspellings

Add deepretro/evaluation: an offline, research-facing harness that scores
ranked single-step precursor predictions with the official ChemCensor engine
and aggregates PT-Max / PT-Top-K plausibility metrics. It is deliberately
separate from the production src/ runtime.

- deepretro/evaluation/: value types, canonicalization (reusing
  utils_molecule), the official-scorer adapter, a SQLite score cache,
  versioned JSONL I/O, PT-Max / PT-Top-K aggregation, JSON report and
  provenance sidecars, and a two-command CLI (score / aggregate).
- deepretro-evaluate console entry point; .gitignore rules for local
  ChemCensor databases, caches, and generated artifacts.
- Tests: core-functionality unit tests plus a self-contained end-to-end
  pipeline test (score via a protocol scorer, then aggregate through the CLI).
  The score CLI path needs the optional ChemCensor engine and is not exercised.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant