Tutorials @ Nesta ✨

A one-stop shop for data analytics tutorials given @ Nesta 🛍️

This repository is to store code for tutorials given @ Nesta. The tutorials are largely from our have-a-go series, a monthly session on a new tool/feature, led by those in the data analytics practice.

have-a-go purpose 🛎️

What’s the problem: DS methods & tools often contain small hurdles to getting started…and documentation is not always great.

have-a-go has started as an informal 45 minute session, once a month on a method / tool / feature / best-practice that a team member (or 2!) has recently learned.

The crucial element of the session is that everyone has a chance to have-a-go during the session, in the way of exercises or break out rooms.

have-a-go process 🧰

So you've decided to have-a-go! The process is a 6 step-er:

1. Decide on what you're keen to learn. Take a look at the existing tutorials to get some inspiration on what's out there.
2. Sign up for a session to recap on what you've learnt.
3. Add code and a high level tl;dr README.md in its own directory in this repo. The readme should minimally summarise what the tool/feature/method is, when its used, why its interesting and perhaps relevant libraries/modules/etc. and links for people to learn more. You can simply name the directory the tool / method you learned. You do NOT need to create a powerpoint for the session, the most important element is exercises/questions that everyone can try to answer in the session.
4. Slack any set up instructions in advance (2-3 days) to the #ds-team channel. Include one sentence on what you'll be demoing and relevant instructions to get started. You can cross-post these instructions to the #data-analytics-chat by reacting to your post with a postal horn 📯.
5. Demo away! You'll know the session is a success if people have a change to use the tool.
6. Once you've done your have-a-go session, please @ the next person in the have-a-go sign up sheet as a reminder. Please also add your demo to the Directory below.

to code review or not? 🤔

Tag someone in a minimal PR to make sure the code runs before your demo.

Directory

Coding

• Code Profiling: Profiling the speed of your Python code with timing tools and Line Profiler ⏱️
• Julia: An intro to the Julia programming language.

Data viz

• Altair Demo: An introduction to plotting with Altair. Includes Nesta styling 💅
• Streamlit Demo: An introduction to creating Streamlit apps.
• Bokeh Demo: An introduction to interactive visualisaitons with bokeh.

Machine Learning

• Reinforcement Learning: An introduction to reinforcement learning.
• SHAP: An introduction to calculating and interpreting SHAP values to for ML explainability.

Simulation

• ABM Demo: An introduction to agent based modelling with Mesa 👾

Data collection

• Collecting data from Twitter API: An introduction to Twitter API v2 recent search a filtered stream endpoints.
• An intro to web scraping: The basics of web scraping.

Data quality

• Data profiling and Data Quality: Using YData-Profiling and Great Expectations to perform EDA, data profiling, validating data batches and for automating data quality checks.

Other

• KNIME Low Code: Reproducing programmes and machine learning pipelines with KNIME, a visual programming tool.
• langchain: An introduction to langchain, a library for building language model applications.
• prodigy: An introduction to prodigy, a tool for creating training data for machine learning models.
• Metaflow: An introduction to building pipelines with Metaflow.