Overall review - C1 W1,2

This is my review for C1 - W1,2 for the MLS.

MLS C1

  • As overall, it is looking great, amazing job explaining things simple for beginners to boost their self-confidence and motivate them to finish the specialization. Good intuitions are delivered about the calculus and algebra concepts for those who are not familiar with it
  • I like the ungraded labs so much. Delivering the programming side of the idea in an easy notebook
    Andrews rhythm in the second week is way better (slower) than the one of the first week
  • The nested hints inside the practice lab are good. Depends on how easy you want it to be for students. I personally think that they should be there anyway because the correction will be available on GitHub after few weeks of the release of the specialization
  • The green cursor does not always indicate the thing that Andrew is trying to explain, and in some occasions the timing of it is incoherent

W1

  • In the “Welcome to the course” video, the slides are entitled “Linear regression with one variable”, which appears in the first seconds of the video. I think this is an issue. Same problem for some of the following videos
  • Andrew’s talking rhythm in the “What is machine learning?” is kinda slow. x1.25 would be great
  • The quiz is very similar to the questions in the videos (maybe even the same)

W2

  • In the video entitled “The model”, at 5:30, the movement of the cursor is not coherent with what Andrew is trying to indicate. He is trying to link a row of the table with one cross in the graph, while the cursor circles the whole table and the whole graph. In the same video “The model”, at 12:00, the cursor appears too early
  • In the “Cost function formula” video quiz, there is a sentence that is missing a word: The cost function measures HOW well the model fits…
  • In the Cost function intuition video, Andrew says that in the next video, we will dive into gradient descent, while the next video is about visualizing the cost function
  • The gradient descent section (three videos + lab) is amazing. I like the simple math implementation and clear visuals. It made perfect sense for me and I asked someone who has no idea about GD to see these videos and explain what he understood. The result was almost perfect. Andrew’s rhythm is also fine
    I only have some doupts about the quizzes in the second and third video of the GD section. IT would be great if you could make sure that the statements are accurate
    The animated visuals in the Feature Scaling lesson are great. An amazing combination between written and popping text. Also the cursor indicated where and when it is needed

Thanks Ammar, are you saying that you prefer Andrew’s week 2 narration because it is slower?

Also, thanks for the details on the cursors and slides. I’ll clean these up for the final version.

Do you recall which quizzes you think have issues?

Yes eddy, week 2 is better.

The quizzes are the second and third lectures of the gradient descent section.