Blog

Title and abstract of Wilkinson et al. (2020), a viewpoint paper published in the Lancet, Digital Health.

Time to reality check the promists of machine learning-powered precision medicine

Reading summary of Wilkinson et al. (2020)
2024-11-04
Zak Varty

Grid of pyramids viewed from above each with one random side coloured in the zvplot colour palette.

Steal like a generative artist

A first forray into generative art.
2024-10-01
Zak Varty

Method of moments for randomised responses, overlaid on an image of a set of scales.

Method of Moments Estimation for Randomised Response Surveys

Method of moments estimation for common data models and randomised response designs.
2023-11-09
Zak Varty

Flow chart showing the stages of preprocessing and analysis used by Potluri et al to quantify data science notebook accessibility.

Notably Inaccessible

A data driven investigation into the (in)accessibility of data science notebooks.
2023-10-03
Zak Varty

BibTeX written on an green background.

Bib but Better

Templates for all the bib entries you never knew you needed.
2023-09-13
Zak Varty

Plot from Chen, Johansson and Sontag (2018). The expected loss for each subgroup of a protected characteristic is shown as power-law function of their sample size. This can be used to inform which subgroups to collect further samples from, and how many further samples to collect.

Why is My Classifier Discrimanatory?

A fresh view on understanding and addressing unfairness in classification (and regression) models.
2023-08-17
Zak Varty

Title page of Bowers and Voors (2016).

How to Improve Your Relationship with Your Future Self

Reproducible workflows are not just good for science.
2023-08-15
Zak Varty

Stylised combination of the CN tower and data icons in purple and orange

Overcoming Barriers to Sharing Code

Talk summary from the Toronto Workshop on Reproducibility in February 2023.
2023-08-08
Zak Varty

Six swimmers in a swimming pool, competing in a race. One swimmer is leading by quite some way.

The Mundanity of Excellence

Exceptional performance is acheived by doing the small, boring stuff really well.
2023-07-26
Zak Varty

Article headline from Real Life - counting the countless, why data science is a profound treat for queer people.

Counting the Countless

Is data science compatible with queerness?
2023-07-14
Zak Varty

Drawing of two people, the first proundly displays a screen full of code while saying “Behold an Amazing Function!”. The second person looks a bit scared and replies “I would rather not behold...”.

Packaging your R code

A tutorial to guide you as you make your first (or hundredth) minimal R package.
2023-01-19
Zak Varty

Two happy looking round fuzzy monsters, each holding a similarly shaped wrench with the word “wrangle” on it. Between their tools is held up a rectangular data table labeled “TIDY.”

Data Wrangling

Get your data ducks in a line using base R and the tidyverse.
2023-01-06
Zak Varty

The text APIs and httr, written on a black hexagon.

Aquiring Data via an API

Part 2 of a mini-series on aquiring data from the web, focusing on data aquisition via APIs.
2022-12-14
Zak Varty

git logo. This is a red-orange square with rounded corners, it is rotated by 45 degrees so that it is standing on one corner and has a cut-out design representing a branching graph.

Git: putting your new feature out into the world.

A git workflow to put a local feature out in the wild, then tidy up after yourself.
2022-12-13
Zak Varty

Hexagonal logo of the rvest package; a sickle cutting wheat on a red background.

Web Scraping with {rvest}

Part 1 of a mini-series on aquiring data from the web, focusing on scraping webpages using the {rvest} package.
2022-12-01
Zak Varty

Header section of a PLOS Comptational Biology perspective paper titled 'Good Enough Practices in Scientific Computing'.

Good Enough Practices in Scientific Computing

Reading Summary of Wilson et al. (2017).
2022-10-19
Zak Varty

BBC dumbbell plot showing stations with the largest gaps between previous and new records, ordered by the highest new temperature.

Data Journalism: Recreating a Professional Plot

Recreating a plot of record breaking temperatures by the BBC data journalism team.
2022-10-15
Zak Varty

Stacked bar chart of yarn weights on Ravelry. Bars are descending order by their fequency, with yarn-type labels written within the bars and count values to the right of the bars. Annotations indicate that the most popular yarn weights are Fingering and Double Knit, and that there were more missing yarn weights than in all remaining categories combined.

Tidy Tuesday: Ravelry Yarn

Tidy Tuesday 2022 || Week 41
2022-10-12
Zak Varty

Photograph of the mathematics section of Blackwell's book shop in Oxford.

Recommended Statistics Books

Book recommendations for those looking to reinforce their knowledge of undergraduate and advanced statistics.
2022-10-10
Zak Varty

Photograph of a filing cabinet filled with library index cards.

Writing a rhetorical précis

Structured summaries to remember and retrieve what you read
2022-10-07
Zak Varty

Mechanical typewriter on white background.

Setting up a quarto blog

Adding a blog within a quarto website
2022-09-26
Zak Varty

The text Hello, World` written in red monospaced font inside curly braces on a grey background.

Hello, World!

A minimal first post
2022-09-26
Zak Varty
No matching items