Conversation
|
@mmcky 👋 This repository doesn't have Copilot instructions. With Copilot instructions, I can understand the repository better, work faster and produce higher quality PRs. I can generate a .github/copilot-instructions.md file for you automatically. Click here to open a pre-filled issue and assign it to me. I'll write the instructions, and then tag you for review. |
Co-authored-by: mmcky <8263752+mmcky@users.noreply.github.com>
|
@copilot when you're finished your work -- can you please resolve the merge conflicts. I update the |
- Updated action versions to actions/checkout@v5 to match main branch - Kept comprehensive workflow improvements with new lecture series - Preserved JAX and PyTorch integration features - Maintained path-based triggering and cross-platform testing
The merge conflicts have been resolved successfully (commit 824a0a2). I've updated the workflows to use the newer action versions from main (actions/checkout@v5) while preserving all the comprehensive workflow improvements including the new lecture series testing, JAX/PyTorch integration, and cross-platform compatibility. |
| push: | ||
| branches: | ||
| - main | ||
| paths: |
There was a problem hiding this comment.
The paths: configuration is a GitHub Actions trigger filter that makes the workflow more efficient. It ensures the workflow only runs when the environment.yml file is modified on pushes to main, rather than running on every push.
Without this filter, the environment testing workflows would run on every commit to main (even for documentation changes or other files), which wastes CI resources. With paths: ['environment.yml'], the workflow only executes when there are actual changes to the environment file that need testing.
This pattern is applied consistently across all three platform workflows (Linux, macOS, Windows) for efficiency while maintaining the scheduled runs for regular validation.
|
@copilot I see the |
Co-authored-by: mmcky <8263752+mmcky@users.noreply.github.com>
Updated the Linux workflow to run weekly on Sundays instead of daily. Changed the cron schedule from |
This PR modernizes and expands the GitHub workflows that test the
environment.ymlfile against QuantEcon lecture series repositories. The workflows now comprehensively test the environment across Linux, macOS, and Windows platforms with updated dependencies and optimized scheduling.Key Changes
Updated Workflow Configuration
actions/checkout@v5,conda-incubator/setup-miniconda@v3, andactions/upload-artifact@v4environment.ymlspecificationenvironment.ymlis modified, reducing unnecessary CI runsOptimized Scheduling
This provides comprehensive weekly coverage across all platforms while reducing CI resource usage compared to the previous daily Linux runs.
Expanded Repository Coverage
Added comprehensive testing for all required lecture series:
lecture-python-programming.myst(existing)lecture-python-intro(newly added)lecture-python-advanced.myst(existing)lecture-python.myst(existing, now with JAX/PyTorch support)lecture-jax(newly added with JAX/PyTorch support)Removed outdated repository
python-lecture-sandpit.mystthat was no longer required.JAX and PyTorch Integration
For repositories requiring advanced ML libraries (
lecture-python.mystandlecture-jax), the workflows now install:jax[cpu])scripts/test-jax-install.pyCross-Platform Consistency
All platforms use consistent bash shells for better cross-platform compatibility and use the same installation procedures.
Testing
This update ensures the QuantEcon environment is thoroughly tested across all supported platforms and lecture series with an efficient weekly schedule, providing confidence in environment compatibility for users across different operating systems while optimizing CI resource usage.
Fixes #6.
💬 Share your feedback on Copilot coding agent for the chance to win a $200 gift card! Click here to start the survey.