Conclusion

With this project, we've once again successfully increased efficiency (or rather, indulged in some laziness).

We hope this solution can help those with similar needs, and we look forward to exploring more automation solutions to optimize daily workflows in the future.

FAQ

1. Why not use GPT-4?

   Because it's expensive. Although the generated content is great, the price is 20 times that of GPT-3.5.

   While we like to be lazy, we can't incur too much expense.

2. Isn't your email content confidential?

   No, these emails are all public. You just need to go to the GitHub page of those open-source projects to see all the content, although we guess you don't have the patience to read through it all.

3. How do you specify which emails to analyze?

   In update_targets_infos.py, you can modify project_names to set the projects you want to analyze.

   # update_targets_infos.py
   # Specify the list of project names
   project_names=("albumentations" "onnxruntime") # <-- Modify this line