This course excellently explained the mathematical and theoretical foundations behind some of the machine learning algorithms, but how to program these algorithms in Python was not explained in the videos and it was left to the viewers to learn themselves in coding assignments, which is the disadvantage of this course. I was just able to do it because I previously had learnt upto some extent from some other websites.

So my advice is that if you still want to take this course, then after learning python, learn python libraries of Pandas, Numpy, Scipy and Matplotlib, and after that learn the sklearn libraries along with some theoretical background, and after that enroll in this course.

The course is good actually. It helps understand the required basics of machine learning. However, I would prefer to recommend this course to people who have basic programming knowledge, since in the labs, it didn't provide any explanation of the code. You need to search and understand the code yourself. I do learn how to use scikit-learn after the course because I study computer science and can read extra documents online. However, sometimes I still feel a bit lost in the code. Hence I don't think that people with no programming background can enjoy the course. To improve the course, I will recommend adding more explanation of the code.

Course provides a good introduction to different machine learning algorithms, how they work and when they can be used. Prior math knowledge will be more helpful in following algorithms and understand each of the algorithms in detail, though it is not necessary since libraries implement the same. Labs were power packed and contain a lot of code that is not covered in this course. Labs assume prior python knowledge with some of libraries used.

Overall i enjoyed the course, but had to look up online to understand some of the concepts explained and also more detailed comments in the labs would have been helpful.

The difference between teaching and taking quizzes and final coding assignment is too big because you make it optional to see the coding in the lectures and in final assignment you give a huge assignment which is technically not equivalent to the teaching process. So, my advice is that please make the lectures more attentive or make the programming exercises more compulsory and more suggestion and hints to understand it better, so that we can actually do the final assignment on our own. I have completed 8 courses of IBM Data Science specialization, believe me I have faced this issue in almost all of them.

The videos are fantastic at explaining the concepts, and all the practical work is in the lab (sometimes there's no overlap in content other than the subject). However, the forum is mostly useless as there are few answers by staff, and a couple answers are links to other forums where you still have to figure out what the answer is among the posted discussions. Some of the labs have broken links or deprecated code. The final assignment is a nightmare, the instructions are very general so while not hard you can get to the final results in multiple ways and therefore peer grading is complicated at best.

Excellent course material and labs, but using IBM Watson for the final project was unacceptable. Watson required multiple attempts at "identity verification" with a credit card, and the permalink that it provided was for an earlier (incomplete) version of the final project. It would be better to have a more robust and simplified system for such a critical part of the course. I would also add that the instructions for the final project could be much better.

The good thing is that you get python scripts that can be the basis for ML with sklearn, and you also get to sort of understand how the different algorithms work. The bad thing is that there is no way that someone that doesn't have prior knowledge with these algorithms can learn about them here. Top, he can learn to be an educated monkey and copy and paste code lines. That might be the goal since that's exactly the challenge in the final project.

Great course for learning ML with Python BUT includes surprisingly many mistakes and typos. Even in the final test there are very misleading copy/paste type of error in the description of the assignment. And many students in the forum have point out those mistakes already two years ago. Not fixing those clear and well reported errors is weird move from the creators and stops me giving more than 3/5 for otherwise superb course.

The parts on regression are previously covered in other courses that are part of the IBM Data Science professional certificate. Overall, there is a lot of information covered in this course but it feels rushed and done in not enough depth. It is an ok course for an overview of machine learning methods, but sits in a weird spot of trying to be too broad while being detailed, but too shallow for a rigorous study of each method.

I understand that this is a higher level course, so it may be designed in such a way to require learners to take bigger leaps, but I did not feel the explanations of what was required on the final were very clear, and once I graded other people's finals, it was clear that it was not clear for almost anyone.

Not a terrible course, the material and the topics were good, but better explanations are needed, I think.

On overall IBM data science professional certificate track: Pros: Content is just good enough, instructors are good. Cons: IBM watson and the platform given to practise on is awful and has terrible performance and reliability issues, most often doesnt work and had an impact on my test deliverables. I personally overcame those issues to some extent with kaggle's and google colab jupyter notebook environments.

There needs to be significant proofreading done on this course by a native English speaker. Additionally, the functionality of IBM tools (Watson Studio, Skills Lab, etc.) leaves quite a lot to be desired. The free tier services and trial accounts often do not work and, thus, you are stuck upgrading to a pay-as-you-go account to finish. The final course project requires some untaught ml skills.

While the course does give a pretty good introduction to the concepts behind most machine learning algorithms and enables us to realize how ML works, the problem lies in the code. None of the code is explained in detail, so the course is extremely theoretical. It basically tells you to copy the code for your own use with small edits but does not explain how to write the code in the first place.

A good course to learn know-how of Machine Learning using Python language so as to facilitate analysis and visualization of data to make effective decisions. I thank the professors to make this course interesting and worth it. Only thing is, videos can be made in a better way so as to facilitate people with non programming background. Maybe some basics of programming would help.

Good introduction, but not complete.

The course does well in introducing Machine Learning, and covers a good range of classification algorithms. However I feel doesn't go the full length. The labs very briefly cover implementation but I find that it falls short. There's a lack of polish in the material, while typos are minor, the labs are can be jarring and hard to follow.

While the videos are very intuitive and helpful, the assignments are lacking in providing machine learning coding skills. Many labs are simply reading others' code, which may be a bit helpful but I doubt if students can come up with the code for other similar problems. I expected better technical skills coming out of hands-on labs and projects from this course.

The course gives a useful insight into machine learning algorithms and model creation using the python library sklearn. I liked the content, even though a little bit more mathematical background would have been nice. The exercises were good, but there was much of it already written in advance for us to use, so I didn't learn as much as I would have liked to.

The course presents a useful overview of basic machine learning techniques without going into mathematical detail. The weekly test questions can be improved to assess the non-qualitative aspects of the topics covered. Using scikit-learn is well illustrated by labs using Jupyter Notebooks. There is plenty of room to update and improve the contents.

Overall, I found this course to be enriching. However, there were more than a few errors and unclear directions in instructions for the final assignment. I hope that the course team is/will update the assignment instructions so that they are in line with current version of the required libraries and will make the instructions more precise.

Very much enjoyed the course and am thankful for the great content, however the peer-grading process created some unnecessary headaches. On how to improve this I posted in the forum here: https://www.coursera.org/learn/machine-learning-with-python/discussions/weeks/6/threads/JmWRnLUqSfClkZy1Kinw6Q

Thank you nonetheless for a great course!

This course provides a great introduction to machine learning. The first 3 weeks are in detail and well explained. The 4th and 5th weeks are not explained as expected. The Labs helped a lot in understanding the practical implementations of the algorithms. However, there should be a little explanation of what is going on in the code.

Material covered is substantial.

You get a good overview of machine learning and some algorithms that are used. (Not always in depth.)

My biggest problem with the module is with the end assigment which is not clear in my opinion (and of some fellow students in the forums who also passed this module) This unclarity is not addressed.

Me gustaron mucho los temas del curso! Los videos son buenísimos para entender la teoría; sin embargo, en los laboratorios no está documentado el código y hay muchísimas funciones nuevas que son usadas y no hay ninguna aclaración de cómo se usan o para qué se usan. Sólo en un laboratorio encontré todo documentado y explicado.

This was by far the most time intensive course, not because the topic is so difficult but because the intructions for the final assignment are so vague and unclear. Otherwise the theory sessions were good. But whole structure of final assignment from definition all the way through marking is not clear and VERY time consuming.

I doubt anybody would learn machine learning from this course. It is more of a refresher course than anything significant. There's a lot of information in the videos, but the notebooks are much more difficult. Many of the visualization aspects of the modeling are already completed in the notebooks with little explanation.