Discover more from The Repo
Hypothesis testing, great style and double decent
Week 43-2023 [Issue 11]
Welcome to The Repo! 🚀
Great work deserves to be shared. Especially in data science.
Each week, I'll curate three gems from the data science community:🗄️ Re: Remarkable Repository 💻 P: Prolific Programmer 🏢 O: Outstanding Organization
I hope you find them as valuable and insightful as I do.
You can find all recommendations in the GitHub repository at finnoh/repo!
Thanks for reading The Repo! Subscribe to support the sharing of great work!
🗄️ Repository | “An R package for tidy statistical inference”
The infer package provides you with a framework for frequentist hypothesis testing. First you specify a hypothesis and then you compare the observed data with data that we would expect under this hypothesis.
I can see this not only as a helpful tool for your own project, but also as a great for explaining hypothesis testing to students in a more applied fashion.
💻 Programmer | Statistics, Data, Science
Michael Clark is a Senior Machine Learning Scientist at Strong Analytics, and has previously obtained a PhD in Psychology.
On his blog he writes about a wide range of topics, from Deep Learning for Tabular Data, to the concept of Double Decent, which refers to an overfitted model’s test error decreasing again when further increasing the model complexity.
Thank you for your work Michael!
The Tidyverse Style Guide
🏢 Organization | “Good coding style is like correct punctuation”
Good coding style is like correct punctuation: you can manage without it, butitsuremakesthingseasiertoread.—tidyverse style guide
On the internet you can find many, many, resources on how to write your code.
Many of them have great ideas, some of them offer a coherent philosophy on how to write any code in any language.
One such resource, albeit with a strong focus on R’s tidyverse framework, is the Tidyverse Style Guide.
I recommend taking a look and adapting some of the suggestions for your next R/Python/Julia/… project.
Do you want more curated content?
Each morning, The Sample sends you one article from a random blog or newsletter that matches up with your interests. Kind of like The Repo, but daily. When you get one you like, you can subscribe to the writer with one click. Sign up here.
The Repo grows through support by The Sample.
Curating the Remarkable in Data Science—Every Wednesday!