This public repository holds the content of Coffee & Coding sessions held at the Department for Education (DfE).
Coffee & Coding is a chance to learn about and discuss coding tools and techniques with guest presenters (coffee optional, but, just between you and me... Tea's better).
Please contact [email protected] if you:
- would like to sign up to the mailing list;
- have an idea for a session; or
- or would like to volunteer to lead a session.
If you would like to contribute to our repo, please read the How To Guide we have created.
Emulating the Department for Transport's (DfT) sensible approach - and to induce standardisation across the different Coffee and Coding programs across UK Government Departments - your folders should adhere to the following conventions:
YYYYMMDD_Topic_Name- All folders should contain a
README.mdwith notes or a brief explanation about the session of that folder
See below for details of past and upcoming sessions.
| Date | Title | Presenter | Abstract |
|---|---|---|---|
| 2019/06/26 | No(-more-)SQL: An applied alternative way to structuring and storing data | Avision Ho | In a no(n)-sequel to a session delivered by former DfE-r, Callum Staff, on hierarchical data formats, Avision Ho will go through why you would use a NoSQL database over a SQL database, how you can set up your own one (for free!) and interrogate it within Python. |
| 2019/05/15 | Learn how to learn R in R | Matthew Dray | You can learn R interactively with the {swirl} package, which prompts you with questions from within R's console. You can install a variety of Swirl courses to learn about anything from basic R to more advanced statistical methods. You can also create and share your own Swirl courses, which is made easier with the {swirlify} package. I'll show you how to do this. |
| 2019/05/01 | Reproducible Analytical Pipeline in Action | Jose Orjales | In this session, Dr Jose Orjales will sprinkle some Spanish spice on a proof of concept (POC) Reproducible Analytical Pipeline. He will use Shadow Measures data - used to show the reforms applied to the 16-19 school and college accountability system in 2016 – to demonstrate how a RAP can be deployed via writing functions. |
| 2019/04/17 | Knitting Machines | Duncan Garmonsway | In this session, I - The Machine - will talk about Govdown, and how to customise RMarkdown with the GOV.UK Design System. I'll demo the new govdown package for rendering documents and websites to use accessible components from the GOV.UK Design System, and show a little of how RMarkdown works under the hood. |
| 2019/03/06 | Automated testing in R using testthat | Peter Curtis | Automated testing means your code is easier to write, easier to improve and easier to review. I will introduce the general principles of a test-driven development workflow and how this will benefit your work. The most commonly used package for testing in R is testthat. I will work through an example where I use testthat in the development of an implementation of a simple greedy algorithm. |
| 2019/02/20 | Gooey Git | David Sands | In this session, David Sands (moi), will demonstrate how to Git it on without a command-line. RStudio provides a Graphical User Interface (GUI) for Git. Git is an astounding version control system that allows your code to be auditable, explanatory, and reproducible. This session will cover how to use this GUI to: commit your code; explore the commit history of other projects; create branches; and, merge those branches together All in the comfort of a cosy R Project in RStudio |
| 2018/12/05 | Using R and Power BI to create impactful dashboards | Tobias Jolly | This session with go through the production of a Power BI app, that helps civil servants plan impactful careers. Please try out the app before the session. On the day I’ll go through the project work flow, focussing on: 1) Web scraping and semi-structured data processing using R; 2) Building a Power BI model; and, 3) Publishing a Power BI report for the whole world to see |
| 2018/11/21 | SQL/Excel to R - How to make the Jump | David Sands | You have likely seen analytical work in DfE being done in R. If you want to enter this world of R, but feel intimidated by it, this talk is for you. We will show you how to quickly and painlessly: 1) Connect R to your SQL database; 2) Pull SQL data into R; 3) Replace SQL statements using dplyr commands; and, 4) Replace presentation of data in Excel with DataTables. So that you can start livin’ it up in R world. |
| 2018/11/14 | Automated text classification/machine learning on the Teacher Workload Challenge 2014 (using Python) | Eugene Chang | I will discuss my attempts to do automated text classification in Python, as part of the 12 week GDS Data Science Accelerator programme. Whilst the analysis was based on free text responses from the 2014 DfE public consultation on teacher workload, I will focus on the techniques and methodologies learnt. Highlights, lowlights, and hopefully some interesting results – all will be revealed! |
| 2018/11/07 | Automating reporting using functional programming | Connor Quinn | I will show off some work-in-progress on a tool to automatically produce summary reports for a school. For this session I will try to show how writing functions in R is similar to SQL at first but very quickly brings very large efficiency gains. Come along if you have heard about this R business but don't have a good sense for how you would start using it in your day-to-day work. |
| 2017/07/19 | Predicting public order incidents on the London Bus Network and Spatial regression in R | Kerry Cella | This session will be in two separate but related parts. Kerry Cella will first present his Data Science Accelerator project which he completed whilst working at Transport for London using their data to try and predict public order incidents on the London bus network. Kerry used relatively novel techniques doing network based analysis and aimed to provide an output that could be used to task police officers. |
| 2017/06/07 | Parliamentary Questions (PQ) tools | Tamsin Forbes | Tamsin presents her Data Science Accelerator project to explore ways of automating Parliamentary Question (PQ) allocation in R; implementing a bit of natural language processing and the random forest algorithm. We'll also look at a tool produced by the MoJ. |
| 2017/05/24 | A Gallery of VBA | Chris Thom | Chris Thom will demonstrate how to use VBA for a variety of tasks, including automated reporting and interactivity. |
| 2017/04/26 | Shiny app | Matthew Gregory | Got an app idea? Here we demonstrate a sensible workflow for building a Shiny app. We use an interactive mapping app as an example. |
| 2017/04/12 | Mapping in R: choropleths | Isi Avbulimen | Limited experience in R but want to create something that looks quite nice? This talk will demonstrate how easy it is to create simple choropleth and Voronoi maps in R. |
| 2017/03/31 | Introduction to the tidyverse | Matthew Dray | Heard about the 'tidyverse' in R but don't know what it is? Matt Dray will tell you more about the tidy style and how the tidyverse packages might make your life easier. |
| Department |
|---|
| Department for Transport |
| Office for National Statistics |
More details on the C & C movement across the Civil Service of the United Kingdom can be found on The Machine's RAP website.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent