The 10 Week Machine Learning Engineering Curriculum

The 10 Week Machine Learning Engineering Curriculum

Now, before I proceed, It must be said that this curriculum was originally adopted by the Paper Club so I take no credit. All I will do is probably add a few adjustments to it.

This curriculum is very hands-on, so do not be expecting complex mathematics to be involved in it, but if you want to gain a deeper understanding I suggest you dive in deeper. However, at an entry-level role you should be fine with using frameworks. Anyway, enough talking(or typing)!

The Curriculum:

  • Chapter 2 End-to-End Machine Learning Project [hands-on]
  • Chapter 3 Classification (precision/recall, multi-class) [hands-on]
  • Text feature extraction (from the scikit-learn docs) [misc]
  • Chapter 4 Training Models (linear/logistic regression, regularization) [hands-on]
  • Kaggle competition submission with linear/logistic regression
  • Advice for Applying Machine Learning [coursera]
  • Chapter 5 SVMs (plus kernels) [hands-on]
  • Chapter 6 Decision Trees (basics) [hands-on]
  • Chapter 7 Ensemble Learning and Random Forests (XGBoost, Random Forests) [hands-on]
  • Chapter 8 Dimensionality Reduction (PCA, t-SNE, LDA) [hands-on]
  • Machine Learning System Design [hands-on]
  • (Google) Best Practices for ML Engineering [misc]
  • Working Production Product using ML

The Curriculum Key:

  • hands-on: Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow
  • coursera: The infamous Coursera Machine Learning course by Andrew Ng
  • misc: miscellaneous(basically anything else besides the two main resources)

Takeaways:

Now, Coursera Machine Learning course by Andrew Ng is by no means a bad course; on the contrary, it’s one of the best courses you can do for getting started with ML. However, the reasons that I only put 1 section inside is because:

  1. The code is run in Octave, which is rarely used in practice
  2. The course does not go over Tree-based methods, which are HUGE in ML today.
  3. The course is very theoretical and lacks practicality; while some people may enjoy that aspect, for me I believe the best way to learn is to do.

Conclusion:

Remember, you can adjust the curriculum as you please, but the most important thing to remember is consistency. you are much better of doing 15 minutes a day every day then only 5 hours one day and none the next.

I hope you found this useful and be sure to stay tuned for more!