In this 3 part series, I’ve tried to distil some lessons to help other data science practitioners avoid some of the most common mistakes made by companies undertaking ML projects. The series will be broken down into lessons about (1) scoping, (2) modelling, and (3) deployment. This is the third of that series.
The ‘Deployment’ phase of a data science project involves actually getting your model into production and in front of real users.
Deployment is tricky and has a lot of different facets. Indeed many of the concepts and tools around deployment are really borrowed from software development. I…
So I’m really interested in the climate tech space and especially how AI/ML can be applied to help make the world sustainable.
In particular I found the Climate Change AI (CCAI) paper has a great taxonomy to start thinking about areas AI/ML applies to the climate tech world. However it does not offer a lot of on-the-ground examples of real startups working in the space. I was also unable to find a really satisfactory list anywhere online so I decided to make my own:
In this 3 part series, I’ve tried to distill some lessons I’ve learned along the way to help data science practitioners avoid some of the most common mistakes made by companies undertaking ML projects. The series is broken down into lessons about (1) scoping, (2) modeling, and (3) deployment. This is the second of that series.
The ‘Modelling’ phase of data science involves the meat of actually building your machine learning model(s). If you’ve nailed the ideation phase, the purpose of the project and the data collection should be clear by now. The modeling phase then, is really all about…
In my work, I’ve become acutely aware of a disconnect between the power and potential of AI research and companies' ability to actually harness these technologies to deliver practical business value.
As a data science practitioner, I come across clients daily who have been burned by failed AI projects and are increasingly skeptical of the ability of machine learning (ML) to deliver tangible value. I’ve helped many of these companies turn their data products around and I like to think I’ve learned a thing or two along the way.
If you’ve ever recorded the results of your Machine Learning experiments in a spreadsheet, MLflow might just be for you!
What is MLflow?
MLflow is a tracking tool to organize and record your Machine Learning (ML) experiments. It is very flexible so you can record anything you want really, but the main idea is to record the parameters of your ML pipeline/model during a run and the metrics/results achieved by that run.
It also has a really nice little Frontend to display results which can be run as a service and looks like this:
I started coding about 3-4 years ago as part of my transition into the world of data science and, honestly, I’ve never looked back. This article is partly a reflection on the joy and flow I have found in writing code, but also on how coding has impacted my way of thinking. I hope it is at least an enjoyable introduction and at most a useful impetus for those thinking about getting into programming for the first time!
Machine Learning | Data Science | Climate Change