Open seanmor5 opened 1 year ago
Making notebooks will be much easier once we have precompiled .so to use so CMAKE doesnt have to be in the Livebook environment. Not a blocker necessarily but just in case someone wants to contribute -- just know that your environment might require some tuning.
Is the lack of precompiled .so
still an issue? If so, perhaps some non-notebook examples would suffice as a workaround?
Is the lack of precompiled
.so
still an issue? If so, perhaps some non-notebook examples would suffice as a workaround?
Hey, thanks for asking! No, that is no longer an issue as each release now has precompiled libraries. Here's the latest release:
https://github.com/acalejos/exgboost/releases/tag/v0.3.2
They will automatically be fetched when you install with Mix.
@acalejos Awesome, thank you!
Would you accept PRs for examples? One I have in mind is for Quantile Regression. The documentation references a worked example:
But I couldn't find one. I ended up getting something working on my end, but I think a worked example would've helped me get there much quicker.
If you are, I was thinking of adapting this example from scikit-learn:
https://scikit-learn.org/stable/auto_examples/ensemble/plot_gradient_boosting_quantile.html
@acalejos Awesome, thank you!
Would you accept PRs for examples? One I have in mind is for Quantile Regression. The documentation references a worked example:
But I couldn't find one. I ended up getting something working on my end, but I think a worked example would've helped me get there much quicker.
If you are, I was thinking of adapting this example from scikit-learn:
https://scikit-learn.org/stable/auto_examples/ensemble/plot_gradient_boosting_quantile.html
PRs are more than welcome! That sounds like a great example to add. Looking forward to seeing it. Let me know if you run into any road blocks.
We should have a number of examples for different types of problems. An interesting notebook is compare and contrasting Exgboost, Axon, and Scholar on the same problems.