Types and Tailcalls

Ultralearning Data Science - Week 8

published on December 16th, 2019

How the Eighth Week Went

This week was the first week where I was completely off. And low and behold, I actually got some things done during this week. I finished the review and summary of the fast.ai introduction to machine learning course, which took way longer than I had thought, but also refreshed many things and led me to sit down and understand a few things I hadn't before. I might even turn one or two of these things into a blogpost. I also got started with the deeplearning.ai and practical deep learning for coders courses. I also started the SVG project by collecting and starting to review some publications with similar goals. And I've read through the first chapter of the neural networks and deep learning book.

I didn't manage to complete the review of the first Andrew Ng's course, so that task is back on my list. This week I would also like to complete the first deeplearning.ai course and go through lessons 2-4 of the fast.ai course all while taking and publishing notes on these courses. In terms of practical experiments, I want to apply the different text classification approaches from the first fast.ai course to my own text-classification problem and also do some practical experiments on the SVG project. I also want to write a neural network and taining from scratch and run it on the MNIST dataset.

Reviewing Goals for Week 8

Looking back at my goals for week 8, here are the results:

  1. [done] Complete review of fast.ai lectures
  2. [failed] Complete review of Andrew Ng's ML course
  3. [done] Start Deeplearning.ai course
  4. [done] Start fast.ai deep learning for coders course
  5. [failed] Apply neural networks on Rossmann challenge
  6. [done] Start SVG conversion project

My Goals for Week 9

I have some free time, but Christmas is comming up and will eat some of that time. Here is what I'm aiming to get done this week:

  1. Complete review of Andrew Ng's ML course, create flashcards
  2. Complete the first deeplearning.ai course with notes and flashcards
  3. Complete lectures 2-4 of the fast.ai course with notes and lectures
  4. Complete NLP experiments on my own text classification project
  5. Do some practical experiments on the SVG project.
  6. Implement Neural Network from scratch and train it on MNIST data.

comments powered by Disqus