Mike Thomas is Chief Data Scientist at Ketchbrook Analytics, with background in credit risk modeling, regression & time series forecasting, machine learning, recommendation engines, and natural language processing. In addition, Mike has extensive data visualization experience across a variety of software products and technologies. Passionate about reproducibility and following healthy software development practices in data science.
Mike Thomas has hosted 77 Episodes.
-
Issue 2023-W37 Highlights
September 13th, 2023 | 37 mins 43 secs
Another adventure with incorporating R packages into a WebR application, annotating your fancy equations in a Quarto PDF document, and unleasing a Bayesian model on UFO sightings data.
-
Issue 2023-W36 Highlights
September 7th, 2023 | 38 mins 35 secs
A batch of R functions to level-up your development tasks, revisting a classic R inferno on object allocation, and a call for proposals to take R's infrastructure to new heights.
-
Issue 2023-W35 Highlights
September 1st, 2023 | 40 mins
The next generation of object-oriented programming in R arrives on CRAN, a novel use of R to automate R scripts and documents for Tidy Tuesday analyses, and a terrific presentation de-mystifying the world of web APIs in R.
-
Issue 2023-W34 Highlights
August 23rd, 2023 | 42 mins 28 secs
A few key practices for data preprocessing leveraging the tidyverse, more amazing wins with open source to process high-dimensional USDA geospatial data sets, and an infinitely fascinating look at how recursion and infinite data structures can be used in your R adventures.
-
Issue 2023-W33 Highlights
August 16th, 2023 | 40 mins 19 secs
Another excellent use case of Nix for solving R package installation woes, a practical dev journey of wrapping C code in an R package, and a guide for using the new refugees R package from UNHCR.
-
Issue 2023-W32 Highlights
August 9th, 2023 | 43 mins 28 secs
How a novel blend of automation and the YouTube API formed a new R-Ladies meetup recording dashboard built entirely with R, the momentum of webR continues with a fantastic guide to create a serverless Shiny app, and a new challenger in the world of high-performance data manipulation libraries arrives.
-
Issue 2023-W31 Highlights
August 3rd, 2023 | 38 mins 35 secs
Reducing usage of for loops with the reduce function from purrr, filling spatial maps with density gradients to account for overplotting, and a fun way to add attribution to your fancy ggplots.
-
Issue 2023-W30 Highlights
July 26th, 2023 | 37 mins 30 secs
How consistent formatting and styling is valuable technique for debugging, a visual tour-de-force of jazzing up your ggplots with the amazing ecosystem of extension packages, and why a little investment in learning HTML and CSS is worth your time as an R programmer.
-
Issue 2023-W29 Highlights
July 19th, 2023 | 46 mins 36 secs
The second edition of the highly-regarded R for Data Science arrives with substantial updates, an adventure with "A Programming Language" that brings new perspectives to functional programming approaches, and a new take on reproducibility in data science combining R with the Nix packaging system.
-
Issue 2023-W26 Highlights
June 28th, 2023 | 33 mins 57 secs
Releasing an Word document table into the land of markdown, a practical overview of sharing your machine learning model with others, and taking local control of checking the builds of your package across computing architectures.
-
Issue 2023-W25 Highlights
June 21st, 2023 | 36 mins 10 secs
Uncovering powerful use cases of the slice() function in the tidyverse, a batch of new features and fixes for column labeling in gt 0.9.0, and a fun journey with tidymodels and visualizations on just how much the Reverend Thomas Bayes may have earned from his own home (probably).
-
Issue 2023-W24 Highlights
June 14th, 2023 | 43 mins 21 secs
A batch of useful patterns for your next R project, the highly-anticipated dashboard components of {bslib} have arrived, and creating circle-based charts with customization using {ggtricks}.
-
Issue 2023-W23 Highlights
June 6th, 2023 | 37 mins 53 secs
Another terrific illustration of open-source collaboration in the latest updates to gptstudio, and a comprehensive journey of web scraping in R to bring much-needed automation to a practical research problem.
-
Issue 2023-W22 Highlights
June 1st, 2023 | 38 mins 38 secs
Another gem in the functional programming toolkit with partial functions, simplifying R package creation using fusen, and a creative visualization of worldwide parliament representation.
-
Issue 2023-W21 Highlights
May 25th, 2023 | 39 mins 6 secs
A must-have resource to get you primed for testing R packages interfacing with the web, how ggblend taps into new compositing functionality for clearer plots, and how R stacks up with Excel in handling dates.
-
Issue 2023-W20 Highlights
May 17th, 2023 | 41 mins 4 secs
Introducing the new ggflowchart package, how a dockerized development environment is another win for reproducibility, and our take on Colin Fay's keynote from the Appsilon Shiny Conference.