tidymodels has long supported parallelizing model fits across CPU cores. A couple of the modeling engines that #rstats#tidymodels supports for gradient boosting—#XGBoost and #LightGBM—have their own tools to parallelize model fits. A new blog post explores whether tidymodels users should use tidymodels' implementation, the engines', or both.
Many hospital systems use machine learning models to help allot limited care resources. A new article on the #rstats#tidymodels website explores claims that these models may be discriminatory:
As data analysts, we know that no model is perfect. Residuals, the differences between observed and predicted values, offer valuable insights into the strengths and weaknesses of our models. Introducing the plot_regression_residuals() function from the tidyAML R package - a game-changer for visualizing regression residuals.
Opportunity Scholars at posit::conf(2024). The application deadline is approaching fast; March 22nd. If you're a strong candidate or know someone who is, please act quickly.
Opportunity Scholarships receive free tickets, a workshop, support for travel and accommodation, plus lots of swag.
(1/2) Introduction to Machine Learning with tidymodels 🚀
This workshop by Dr. Nicola Rennie @nrennie provides a great intro to the foundation of machine learning with R examples using the tidymodels framework. The two-hour workshop covers the following topics:
✅ Foundation of ML (training approaches, hyperparameter tuning, etc.)
✅ LASSO regression
✅ Random forests regression and classification models
✅ Support Vector Machines
The latest release of the tidyAML R package improves access to model predictions.
The internal_make_wflw_predictions() function now returns a single tibble containing the original data, training predictions, and testing predictions. This makes it easy to evaluate model performance without extra wrangling to extract the predictions.
Dive into the latest tidyAML release! 🌟 The spotlight is on the new .drop_na parameter, enhancing the functionality of fast_classification() and fast_regression() functions.
We had an amazing talk yesterday from Emil Hvitfeldt on What's New In Tidymodels where we learned about some handy new data processing functions, new capabilities in survival / time-to-event modeling, and the nascent projects on causal inference and fairness coming soon to the tidymodels ecosystem!
Upcoming event!
What's New In Tidymodels with @emilhvitfeldt
The @RUGatHDSI will be hosting this event on Thursday at 5pm Eastern Time.
"The tidymodels framework is a collection of packages for modeling and machine learning using tidyverse principles. This talk will touch on a number of new additions and in-process work being done by the team."
A friendly reminder that this event is coming up tomorrow at 5pm Eastern Time! Register on our website here: https://rug-at-hdsi.org/calendar/
We know for a fact that Emil has been up to some great stuff recently (see his recent blogpost on tips for Quarto slide-crafting here: https://emilhvitfeldt.com/post/slidecraft-7-tips-and-tricks/) so we're anticipating this will be a great talk!
Soon I'll buy my Super-Fan tickets for #PositConf2024 in Seattle (not available quite yet as far as I can find), but first it's time for one more thread to summarize my threads! Each post in this thread will be flagged with a titled "content warning" to make it easier to find your way back to the top, I hope that works out!
The workshops were a wonderful new experience. I TAed the #TidyModels workshop last year, but I'd never actually participated in a Posit/RStudio workshop. There will only be 1 day of workshops next year, but I definitely recommend finding one to participate in!
Bringing some super limited edition #tidymodels hex stickers with me to #positconf2023. Check out my talk Tuesday the 19th at 2:40 in Ballroom CD to find out what’s up with the T. rex and LEGO bricks.