• What we'll do
We'll have two talks from Thomas J Fan and Joseph Cardenas
Thomas will give his talk "Deep Dive into scikit-learn's Hist GradientBoosting Classifier and Regressor"
Gradient boosting decision trees (GBDT) is a powerful machine-learning technique known for its high predictive power with heterogeneous data. In this talk, we will explore scikit-learn's implementation of histogram-based GBDT called HistGradientBoostingClassifier/Regressor and how it compares to other GBDT libraries such as XGBoost, CatBoost, and LightGBM.
Thomas J Fan is a scikit-learn core developer working as a Research Associate at Columbia University's Data Science Institute. In his free time, he maintains skorch, a scikit-learn compatible neural network library that wraps PyTorch. Prior to his work in open source, Thomas studied to be a theoretical physicist at Stony Brook University and NYU.
Joseph will give his talk WebAssembly for Pythonistas
A general introduction to WebAssembly ("Wasm") - what it is, how it works, and why Python programmers ought to care. The talk will go over the above and include a demo written in Rust. The latter is being used because there are currently fewer tools to get Python compiled to WebAssembly vs. Rust/C/C++.
• What to bring
Your wonderful self.
• Important to know
PuPPy has a Code of Conduct that all attendees are expected to follow. The Code of Conduct can be found at https://www.pspython.com/pages/code-of-conduct/