Translating from {tidymodels} and scikit-learn: Lessons from a 'bilingual' course
summary
A bilingual course that duplicates exercises in tidymodels and scikit-learn reveals concrete mismatches: selecting a single penguins column yields a one‑dimensional vector in R but a two‑dimensional dataframe in Python; SVM tuning parameters invert meaning (R’s cost vs Python’s C); k‑means defaults like Hartigan‑Wong versus k‑means++ choose different initializations so outputs rarely match. Students using R emphasize feature selection, Python students emphasize preprocessing, and confusion persists unless defaults, parameterizations, and shortcut helpers are documented and translated.
AI-written summary · about
chapters (5)
- 0:00Start
- 10:18"FALSE FRIENDS"
- 11:04SUPPORT VECTOR MACHINES
- 13:54SLANG TERMS
- 16:38NONVERBAL COMMUNICATION
related talks
57mGood Practices for Applied Machine Learning
01h 24mInpsect and try to interpret your scikit-learn machine-learning models
21mKeynote: Easier Machine Learning Thoughts From Scikit-Learn
39mBeyond Bias: Crafting Fairness in Pythonic Machine Learning Systems
33mUsing Scikit-Learn's interface for turning Spaghetti Data Science into Maintainable Software
15mInterpretable Machine Learning with R
01h 17mMachine Learning with Scikit-Learn
30mKick-Start your Understanding of Machine Learning with Python