Netflix or Coursera? How to finish Andrew Ng’s 1st Deep Learning Course in 7 days

Jan Zawadzki
Machine Learning World
4 min readNov 8, 2017
You will have to stay idle for a week now

If you love Andrew Ng’s first Coursera course on machine learning as much as I do, you were equally hyped when you heard that deeplearning.ai finally released the follow-up lectures about Deep Learning.

Since everybody’s on a tight schedule, let’s try the impossible and finish a course that is laid out to last one month in one week. Let’s not rush through though, but actually understand the material. And of course, we’ll do it while continuing our 40h/week job.

What are the advantages of finishing the course quickly you ask? Well, you can enjoy the entire course during the free enrollment plan of seven days, you can start the next course earlier, you save money because you pay a recurring subscription fee per month, you really have the chance to understand the material better when it’s fresh on your mind etc. etc. etc. …

We stated the goal — Finish Andrew Ng’s 1st Deep Learning course in 7 days — how do we get there? Let’s use this project plan I created so all we have to do is postpone the binge watching session of The Walking Dead and execute the plan.

Let’s begin the challenge.

A few tipps before you get going

  1. Work when your mind is fit and healthy — there is no need to force your brain to understand deep neural networks when you’re tired, so try to study when you’re feeling energetic and ready to learn
  2. Watch at 1.5x speed — Andrew is so kind to speak slowly for non-native english speakers but your mind will adapt quickly if you increase the speed. The project plan relies on watching the lectures at increased speed — after all, this way you can save more than 2h of your time!
  3. Snack before you study — eat nuts, veggies, fruit or a healthy dinner before starting to study so you have the energy you need
  4. Conduct your work in one sitting — short bathroom breakes are ok, but don’t get distracted swerving on Facebook while grandmaster Ng is trying to teach you cool stuff
  5. Don’t overcomplicate — 90% of the answers are actually right in the hints and comments of the code, so don’t try to come up with new approaches. Check the forum after you’ve been stuck on an exercise for more than 10 minutes and never hesitate to reach out for help

The project plan

The course is divided into four weeks and three programming exercises.

Project plan to structure 7 days of deep learning

The project plan shows how much time on which day you should allocate to a given task. The project plan starts on Wednesday but feel free to start at any given day of the week. Nevertheless I would recommend starting Monday — Wednesday, so that you have the weekend to work on the week three and four assignments.

Start off by watching all the lectures for week one and finish the test right away. It should not take you longer than one hour and will get you excited for the upcoming exercises.

The lectures in week two will take longer than in other weeks. If you’re familiar with numpy and python broadcasting, you can safely skip the second part of the video lectures and will still be able to complete the programming exercise. Try to spend half an hour on the optional numpy programming and 1.5h on the logistic regression exercise. If you read the exercises carefully, you should be able to finish the tasks in two hours.

In week three things will start to get interesting. If you follow the project plan, it should be Saturday now. Finish the lectures plus the quiz in one hour and the programming exercise afterwards in two hours. This will take up half of your day, but afterwards you still have half a day left.

Week four will prove to be the most challenging part. If possible, try to finish the lectures and coding exercises on Sunday. But to keep you mentally sane, it is also no problem if you just watch the video lectures on Sunday and you take Monday and Tuesday to finish the coding exercises.

Et voilá — you pulled through and finished an online course in one week what should’ve taken you a month!

Now is really the time to be proud of yourself, pat yourself on the shoulder and check if the living dead are still walking and dead.

Thanks for reading and if you have any questions about the project plan or the course, feel free to reach out to me.

--

--

Jan Zawadzki
Machine Learning World

CTO & Co-MD CertifAI | Focusing on reliable AI engineering to spearhead the next wave of AI adoption