The course is good but I expected a faster response for my regarding assignments & course materials .

Can you guys make sure that your mentors reply faster to student queries ?

You need to make assignments more descriptive as lot of time is being spent on forums to just understand the problem correctly.

The autograder behaves erratically lot of times so you need to make it more efficient.

Nothing wrong really. Should have provided more mathematical theory in the resources section.

Assignments should be a lot tougher and on real life data sets which require recodings and transformations. Quizzes should be more relevant to the lessons taught. More hardcore theoretical resources, like books and research papers should be included in order to complement the practical lessons.

This course was too theoretical and lacked any practical exercises that would help me solve any problems. The professor went too deep into the concept and in the end you were left wondering what is the purpose of the algorithm. Seems as if they were concerned in covering a specific amount of topics rather than making the concept of machine learning more approachable.

Hi

First I want to thank all the instructors and anybody that was involved in this course preparation. That was a great opportunity and I really liked that but not in all parts . I think the syllabus was a little heavy and somehow I couldn't follow that . in the programming part I needed more guide and sample .

But in general It was good and I thank you so much.

I struggled with this course. The lectures cover a great deal of information extremely fast. I appreciate that there are more lectures than in previous courses in the specialization and the information is better presented IMHO. The assignments were quite difficult and I struggled. Relying heavily on discussion forums and online posts.

I think the instructor must give more practical explanation for scikit-learn. I need to research almost everything for completing a particular assignment. Please have changes in pitch of your voice while delivering the lectures so the lectures don't seem boring. Also, please update the autograder !

Overall a good course. Thank you.

The course material is good, but the teaching style is too boring. Without the standstill slides, if there is animation, it would be helpful for us. And, the assignments are not straight-forward and the autograder is buggy. As an example, I can run the assignments easily in the jupyter, but the autograder faces problems.

There are too many errors in the video and even in the quizzes and assignments which will affect the final grade and wastes studying time to figure out it is an error. It is pointed out in the discussion forums but no one is taking the action to correct it. Moreover, at least 3 of the reading materials fail to be loaded.

Good course for beginners. However, things like feature selection, dealing with null values, model selection should be in depth and an end to end example on a real world dataset should be explained step by step to with best practices to develop learner's interest towards picking up problems and solving on their own.

Course is good but leaves a lot of things unexplained and feels like the weeks explaining ml algorithms are in a rush.But the assignments are truly remarkable.I would recommend this course to anyone who already knows machine learning and would want to apply it on some good problems/assignments before Kaggle.

I really like the main idea of this course, i.e., using sklearn lib along with basic lectures on the ML topic. So, I was expecting that we will be following the contents of text book by A.C. Müller & S. Guido. In the first two weeks it was really good. The materials of last two weeks were quite compressed.

Its a really complex topic an though videos seem long enough to explain some ascpetcs of it, many little things go under the radar and make it difficult to understand some thing. Algo, the lectures are a bit weird since the professor sometimes stutter or changes ideas mid sentence.

Good explanations on videos, The only problem which was really time consuming and wasting was the problems related with the assignments submission. but overall this course helped me a lot to structure machine learning fundamentals in my mind and to get a good practice out of it.

The videos are way too long and very monotonous. They should be cut down and reduced. The maximum length they should be is 5-6 mins other wise they becoming distracting.

The course content is good though. Quite easy to understand but going through the videos is a chore.

I think that the subjects are very advanced. There should be a more clear specifications of prerequisites for the course. I had to look for lot of help outside the materials provided for doing the activities. The course is fine if you have the apropiate skils though.

It's a good course, but a quick one. One needs to have a beforehand knowledge of all the algorithms as they are not discussed in details. State of the art is not mentioned. Implementation and best practices are present, along with pros and cons of each algorithm

The course is quite high-level. There is nothing wrong with an applied course being high-level. The material is easy to follow, the quiz is a bit challenging but the homework assignments are quite easy to pass. I prefer a course with more fundamental details.

The course is too tight, just 4 weeks cannot master the machine learning. This course can splitted into 2, in order to capture more on the deep learning and unsupervised learning, which are important, but being categorized as option in the course.

the second assignment was a little beyond what was taught in the lecture. others are fine.

big suggestion: please please have a better auto-grader. Most of my time was spending on how to battle the auto-grader instead of coding...

Course is well structured, course material also is well defined and learning is excellent. Though Instructor's communication is very laidback. Should have more engagement in tone and connect with enthusiasm.

Quick tutorial-like overview. Autograder is not too verbose and as a result I spent some time struggling with debugging the code rather than figuring out how to solve machine learning related problems.

Good content and I learned a lot. However, the instructor made too many mistakes during the lectures and the assignments also have mistakes that need to be fixed by the students.

Follow the course with introduction to ML with python to have descent understanding. Instructor won't be able to keep one interested for long. Exercises could have been tougher.

Two point: 1) you can learn a lot here, 2) imagine you are shown a hammer but never explained how to hit a nail. Two previous courses in the specialization do both.