Conversation
- ci.yml runs on every pull_request event and pushes to main: build and test with default traits, benchmark-leakage check, strict dataset validation, plus a separate CoreML+NaturalLanguage trait build job. SwiftPM build directories are cached per trait flavor; in-progress runs for a ref are cancelled on new commits. - evals.yml is workflow_dispatch-only: runs the extraction-affected eval suites under coreml_default, gates the pressure and smoke baselines, and uploads run reports. Manual because eval results are OS-sensitive; the release gate stays local per AGENTS.md.
Member
Author
|
Superseded: CI moves onto the linguistic-tagging branch. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
ci.ymlruns on every PR commit (opened/synchronize/reopened) and on pushes tomain:swift testwith default traits, thenScripts/check_benchmark_leakage.pyandmemory_eval validate-datasets --strict.swift build --traits CoreMLEmbedding,MemoryNaturalLanguage, catching trait-gated compile breaks that plainswift testmisses..buildcached per trait flavor (seeded bymainpushes); concurrent runs on the same ref cancel in progress.evals.ymlis manual-only (workflow_dispatch): runs the five extraction-affected suites undercoreml_default, gates the pressure and smoke baselines, and uploads the run reports as artifacts. It is deliberately not PR-blocking because NaturalLanguage model behavior varies by macOS version; the authoritative release gate stays local per AGENTS.md.Notes
macos-15with the newest installed Xcode selected. If the runner's NaturalLanguage models diverge from dev machines, the NL-dependent unit tests will surface it on this PR's own run — adjust from evidence if so.Testing