tldr The course is a great introduction to ML for an audience already comfortable with mathematics and Python. For what it aims to achieve, I think it does a great job. /tldr

The mathematics involved in the first course of this specialisation is not that difficult if you already have a solid foundation on calculus. Some functions used in the Optional Labs are called for you from already written python scripts (which you have access to, and can download to inspect). The first 3 weeks (and probably the rest of the course) will not teach you fundamentals on Python or mathematics or statistics, and some details regarding the choice of loss function for logistic regression were omitted. Furthermore, libraries such as scikit-learn were used to complement the material, but not explained in depth. (Granted, this course is not about Python libraries.)

All in all this seems like a great introduction to ML for people already comfortable with mathematics and Python.

If you already have the foundations required (Undergrad basic calculus, Python) you can do all 3 weeks in one day fairly easily without distractions.

It is the Best Course for Supervised Machine Learning!

Andrew Ng Sir has been like always has such important & difficult concepts of Supervised ML with such ease and great examples, Just amazing!

In general, I think it was a valuable course to take. I like the way Andrew tried to conveying the ideas intuitively to make sure the students understood the methods behind the learning algorithms. However, I would've loved if there was more in-depth treatment for the Math aspects of the obtained results. Also, the assignments + Optional labs were not as engaging as I hoped. What I mean by that is, it almost required no deep thought from our side to implement the procedures. In other words, there was a lot of skeleton code that makes you "implement" the algorithms with almost no thought (which I don't think is beneficial to the student's learning experience)

It's completely fine. I have learned a lots of thing in this first course of specialization. Thanks to courseera for giving such a good and fine course on financial aid. I am very thankful to them.

Excellent content. I'm a math guy so I would have enjoyed some more in-depth theory, but that's what books are for I suppose!

I've been using Python for a long time now so understanding the code was nice and easy.

Thank you for your hard work putting this together!

I started with onld ML course last year, completed successfuly but did not purchase the certificate. As I am more familiar with python than Octave, this new course make thing clearer for me.

The explanation is clear, and all of the source codes provided in each jupyter notebook show a clear visualisation of how well the model learns or fits into the data when a parameter changes.

The quizes are too straight forward and simple. The code exercise too short as well.

Also disappointed that vectorisation is introduced but cost and loss functions are still calculated in for loops.

Teaching is an art and Andrew Ng is a great artist. He explained everything in the course in the details and with examples easy to comprehend. Thanks a lot for helping thousands of students like me.

Professor Andrew can explain complex knowledge clearly. The Python lab can help learner to understand algorithm. The course is more valuable. I am excited to learn the next course for advanced ML.

Excellent balance of theory and practice provided by exceptionally well documented and visualized examples and code in Jupyter Notebooks that one can interact with to build intuition.

I have just finished the old machine learning course, and I'm doing this because I'm learning python/numpy/matplotlib. I thought the question during the course and quizzes insulted my intelligence. The material is great, but you need to improve the simple questions and quizzes. The first programming assignment was too easy, the second programming assignment was at a fair level. I still think more should be left to the student to do.

This course is helped me a lot . I gained some skills related to the supervised learning .this skills i learned in this course is very helpful to my future projects , thank u coursera and andrew ng

Andrew Ng is the best proctor for Machine Learning. The course has been perfectly balanced with thoritical as well as practical aspects. After this course I feel so confident. From ZERO to HERO

absolutely amazing course, coding assignments are designed perfectly and the course helps in understanding the working and the math behind the algorithms which makes it so recommendable.

a great course to understand theory of supervised machine learning. Need lectures for numpy and scikitlearn

The best of the best. I am superglad to see the upgraded version of the legacy Machine Learning Course by the super helpful tutor, Andrew Ng, implemented in Python. Very detailed Labs, allowing plenty of practice and intruition. Luckily enough, I was already great at Python and NumPy. I hope the Labs won't be intimidating to a Python beginner.

Overall, this course deserves more than 5 stars. It is second to none, as far as my exposure to Machine Learning is concerned. Thanks Deeplearning.AI and Standford for creating such a fantastic course. I am definitely taking the remaining courses in the specialization😊

Excellent course. Intended as a refresher, and had a better understanding of feauture engineering, scaling, and logistic regression. Good hands on labs were very practical, engaging and rewarding.

Really enjoyed the course, had a few questions by the end of it that were resolved quickly in the forums. I would implore others to use them too as they are a great resource.

Andrew Ng is a very good professor, he explains complex concepts in a very simple way and with the help of many visualization and graphing tools. Highly recommended course!

The course is very interesting. I have learnt a deep understanding on machine learning, now I know the difference between regression and classification.

Good, I keept working on these codes and searching for clues in videos. Good structure, reinforcment of some knowledge.

Andrew Ng is a God of ML. No one in this world can make this course more easier than him.

Pretty good introductory course! Personally, I would like to see more time devoted to the Scikit-Learn implementation (and maybe Pandas data frames instead of NumPy arrays for the training data) as opposed to hard-coding the algorithms and using really small data sets. Scaling upwards and using those libraries on larger data sets should be relatively easy after you nail the foundational concepts in this course, though. There is definitely something to be said about knowing the mathematical algorithms running in the background of these black box models, and this course does a really good job of explaining them (namely, cost functions and gradient descent).

Apart from scripting these algorithms in Python code, the course is somewhat lacking when it comes to conceptually explaining regression and classification models. For example, there is no time spent explaining how to interpret regression model coefficients and intercepts, and there is little time spent explaining the probabilistic interpretation of the sigmoid function and the importance of choosing a good decision boundary. It is one thing to know how to program these models and another thing to be able to explain them to people without a technical background, which I think could be a good lesson in future versions of the course.

Overall, great introduction to the models and their implementation in Python! I would absolutely recommend the "optional" labs throughout the course (especially if you're new to Python) because they show you the code that you'll have to write in the required assignments.

if labs were optional then why are there compulsory coding assignments, labs must not be optional, instead make us type code step by step, like MATLAB onramp courses.