It was good untill week 3. My real point is 3.5 . I think this spec is definitely not a math course. they just show the math parts of the ML , they are just telling ML terms, this is a calculus course but subjects are what a neural network is, what a gradient descent is , or network method etc. Where exactly math here? There is no need if you dont tell it comrehensively , Every tutor is teaching much more math than this spec in any ML course. You are telling %10 percent math and then %90 percent ML terminology. Why do i have to learn what a neural network even with more than two layer in a math course. Then what are you gonna tell me in the ML course??? if you will, then make it like first two weeks. It was like equally math and ML but in week 3 it is %90 like an ML Course.

Overall, it is an excellent introduction to calculus and applications to ML. However, some of the foundational explanations of differentiation need clarification. If you want a rigorous introduction to calculus, I recommend Khan's Academy. For example: in Khan's, you will find an in-depth explanation of limits which is only touched upon briefly in this course.

Rather simple introduction to Calculus in machine learning & data science. The course covers core concepts and linked well to applications such as Optimisation and Gradient descent.

The instructor provided good graphical visualisations to help learners understand and develop intuition on the concepts covered.

I actually understand gradient descent which is awesome. I need a little bit more practice running some of the problems to be proficient and remember how to do them, but I was able to complete them for the course and I suppose in real life I can just have my code (stolen from this course) find the answers.

An excellent course indeed, although with some caveats.

I love the simplicity and the examples of the instructor. However, sometimes the materials needed to be more complex and exciting to watch. I definitely recommend this course to absolute beginners in calculus or someone who needs a refresher.

Content was extremely helpful but assignments were too simple. If you could make the graded content a little more challenging it might really help the student push himself further.

We had a great time learning this course. We really had some good sessions with friends while learnin this. We found about new concepts. Thanks

Easy to follow for beginners. Concepts are well explained. I wish the Newtonian method had been explained in more details though.

The last programming assignment has some problems. If you touch the optional part, you get 0.

Very good course in terms of content. Labs and assignment env was a bit flaky when running.

It was so hard and challenging. I've nearly cried and somehow i passed. Thanks :")

Too easy for exercise, but the video lession is good, focused on ML perspective

It great course but the numpy will be new for some one who don't know to code.

Labs could have been more difficult. Otherwise a good course for beginners.

Videos and explanations are great, but labs are rough.

Helpfully and covered all the topics related to ML

Good course and gives a good understanding!

Notebook test case sometimes crashes

You are perfect Platform Coursera

Nice for review.

Good Professor !

kind of tuff

The material was great and very informative, and intuitive. That said, there were a few times when 1) the lab didn't align with the learnings - multiple variables were used in labs before described in later lectures. 2) The final quiz asked questions about using Newton's method approximation recursive formula, but the language used was very different than how the concept was taught in previous lessons. If mentors/contributors can look through the lessons/labs/quiz and ensure there is fluidity, that would be very helpful. Thank you!

Much time spent on Python in the labs that did not result in learning Python.

The labs make it clear that Python is VERY important, but I suggest a separate Python course to be taken first (even before the linear algebra material). Even making it into a four-course specialization is a far better use of one's time than struggling with labs that only8 illustrate, not teach, this critical tool.