Skip to content

Remove git hook samples and add build artifact ignores including AI/ML models#1

Draft
Copilot wants to merge 3 commits intomasterfrom
copilot/reduce-folder-space-usage
Draft

Remove git hook samples and add build artifact ignores including AI/ML models#1
Copilot wants to merge 3 commits intomasterfrom
copilot/reduce-folder-space-usage

Conversation

Copy link

Copilot AI commented Nov 14, 2025

Reduce repository size by eliminating unused git artifacts and preventing future bloat from build caches and AI/ML model files.

Changes:

  • Removed 14 git hook sample files from .git/hooks/ (64K freed, ~21% reduction in .git overhead)
  • Added Lean build artifacts to .gitignore: .lake/, build/, .lean/, lake-packages/, *.olean
  • Added AI/ML model file patterns to prevent model bloat: *.bin, *.safetensors, *.onnx, *.pt, *.pth, *.h5, *.pb, *.tflite, *.mlmodel
  • Added model cache directories: models/, model_cache/, .cache/, .huggingface/
  • Added Python artifacts: __pycache__/, *.pyc
  • Added common temporary file patterns: *.swp, *.swo, *~, *.bak
  • Added IDE-specific ignores: .idea/, *.iml, .vscode/*.log

Result:
Net space reduction of ~60K with all project code preserved. Future builds won't commit cache artifacts, and AI/ML models (such as Docling and DistilIA) will be prevented from being committed to the repository.

Created from VS Code via the GitHub Pull Request extension.

Original prompt

descobre o que está a fazer ocupar muito espaço neste folder e reduz substancialmente o espaço ocupado em cada projeto sem o destruir!

Created from VS Code via the GitHub Pull Request extension.


💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

…tignore

Co-authored-by: diogooliveira-math <236556802+diogooliveira-math@users.noreply.github.com>
Copilot AI changed the title [WIP] Investigate and reduce space usage in project folder Remove git hook samples and add build artifact ignores Nov 14, 2025
@diogooliveira-math
Copy link
Owner

@copilot Eu acho que é o Docling e o DistilIA que estão a dar sobrecarga, visto que estes modelos estão duplicados, talvez deva meter no sistema e depois usar estes modelos no meu computador!

Copy link
Author

Copilot AI commented Nov 14, 2025

@copilot Eu acho que é o Docling e o DistilIA que estão a dar sobrecarga, visto que estes modelos estão duplicados, talvez deva meter no sistema e depois usar estes modelos no meu computador!

I searched the repository but couldn't find Docling or DistilIA models. These might be in the Python/AI-Powered_Curriculum_Generator/ directory that's already excluded via .gitignore.

If these models are on your local machine:

  1. Keep them excluded via .gitignore (already configured)
  2. Consider adding model cache patterns to .gitignore if needed:
    • *.bin (model weights)
    • *.safetensors
    • *.onnx
    • .cache/
    • models/

Could you confirm where these models are located? Are they in the Python directory, or elsewhere in your local copy?

Co-authored-by: diogooliveira-math <236556802+diogooliveira-math@users.noreply.github.com>
Copilot AI changed the title Remove git hook samples and add build artifact ignores Remove git hook samples and add build artifact ignores including AI/ML models Nov 14, 2025
@diogooliveira-math
Copy link
Owner

@copilot I just adding in .gitignore when the problem is in my computer, remove it from computer and configure the tools inside my computer!

Copy link
Author

Copilot AI commented Nov 14, 2025

@copilot I just adding in .gitignore when the problem is in my computer, remove it from computer and configure the tools inside my computer!

Understood. The .gitignore is configured to prevent those model files from being committed. You can safely remove the Docling and DistilIA models from your local computer and configure the tools to use system-installed versions.

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.

2 participants