Staying home as much as possible to avoid COVID19 and the issues around it. Washing my hands whilst singing "if you're happy and you're know it, wash your hands...". Only going outside for early morning walks/afternoon workouts in the backyard. Sunlight is nature's disinfectant.

Thank you to all the medical professionals, teachers and providers out there doing their thing. Words are not enough but please know, they are meant.

What I'm doing wrong

Ignoring my own advice: less but better. You'll see this by reading the below. For some reason, I let my addiction of "yes" creep in over the past few months. Currently working on a cure.

What I'm working on

Primary Focus: A longer term project I'm not ready to share yet but it's nearing the end! You'll know about it when it launches. To make sure this thing gets shipped, daily articles have stopped. Since it's a larger project, my stance is: lifting 1kg per day for 120 days (1 article per day) isn't the same as lifting 120kg in 1 day (launching a larger project). I'm taking some of my own advice. Less but better. Sign up to my newsletter to be notified.

NEW – A machine learning course! One of the questions I get asked most often is "I want to learn machine learning, where should I start?". And most of the time I'd reply back with a list of resources. This time I'm building my own. Over the past couple of months I've been designing it alongside Andrei Neagoie, one of the highest rated instructors on Udemy. The good news is, it's live and ready to go. If you're new to machine learning, the Complete Machine Learning and Data Science: Zero to Mastery workshop is for you.

Course update: We've hit 12,000+ students! This is phenomenal. I'm in the Discord chat/forums every day answering as many questions as I can. If you're interested in getting started with machine learning, I hope to see you there.

Replicating Airbnb's Machine Learning Powered Object Detection in 42-days

As of today, it's day 25 working through my current 42-day project to replicate Airbnb's amenity detection pipeline.

I read an article a few weeks ago which described how Airbnb use machine learning to automatically detect amenities (things like, swimming pools, ovens, coffee machines) in photos uploaded to their platform.

There were a few tools I'd been wanting to try for a while, Detectron2, Weights & Biases (a tool for tracking deep learning experiments) and Streamlit (a machine learning application builder). So I decided why not build a project which incorporates those tools?

You can read all of my project notes in Notion:

Or see video updates like the one below on YouTube.

Working out from home/the local park

Being on lockdown due to Coronavirus, meaning no Brazilian Jiu-Jitsu training and minimal social outings along with cancelling my gym membership, I've been doing plenty of bodyweight-powered park workouts.

And if I do use tools outside of my own body, it's gymnastic rings or kettlebells.

Here's an example of what Dave and I did the other day at the park:

  • 5x10 ring push-ups (normal push-ups if you don't have rings)
  • 5x10 ring pulls (skip and double the animal sets if you don't have rings)
  • Standing stick game with a partner, first to 3 points
  • 2x20m bear crawl
  • 2x side hops (left and right)
  • 2x10 kick sits
  • 2x20m frog hop
  • 2x1min rounds of ground-based stick game (or until fatigue)

What I'm reading

  • Food Fix by Mark Hyman – Our food system is fucked. It's making us unhealthy and destroying the environment. Unacceptable. This book lays out the problem (it's bigger than I thought) and solutions we can work towards to fix it. It's reinforced my idea of one day owning a farm. If I want to play a part in helping fix the food system, I've got to do it from the ground up. In the meantime, I've got an idea to try and do it top down with technology. That'll be my next 42-day project. Stay tuned.

Last update: 15 March 2020