Fairly okay course. Lectures were real hurried and high level. Had it not been for my Math and CS background I would not have gotten most of the material.

Several lectures were superficial. I feel like I need to put a lot more time on my on to understand how to build a proper neural network model.

The programming assignments were easy. I wish it had some more difficult assignments requiring you to code your own neural networks

Good course, the explanation of CNN could be better, the anomaly detection exercise it's kind out of focus

Basic Deep learning; no programming assignment for most part of deep learning part

Os vídeo disponívei no youtube estavam tinha apenas a opção de legenda em alemão

Hard to follow ... found a lot of assistance in discussion forums

The concepts were rushed without concise explanations.

unfortunately the time series chapter is not complete

The lessons provided by Skymind were very poor.

Assignments are a bit too simple

I liked the general idea of this course, but the actual material is not as good as it could be. There are lots of inaccuracies in the material (like annoying typos and not working code examples) which should be corrected before you sell this course on Coursera.

I strongly suggest that you go through your material with someone who has pedagogy knowledge and who can assist you to improve the didactic aspects of your material.

I did this course (and the whole specialization) for the practical examples as I feel rather confident with the theoretical aspects of machine learning, but I wanted to learn how to do these things in Spark environment. At the end of the day I have got what I wanted (more or less, as the NLP part was really lousy), but if I would not have strong experience with the field, I would have been surely lost. Honestly, I would have a hard time to recommend these courses for someone who wants to learn about machine learning and not about how to do machine learning with Keras, etc. And I am sorry to say that, because, again, I liked the team, the attitude, and the technical aspects of this course.

There was a lot of interesting content, but I was sad that the programming assignments were fairly trivial. Any point where something deep and useful could have been assigned (I was hoping to get experience or guidance on building an LSTM autoencoder, for example) we were instead given a super easy alternative that was mostly pressing [Shift]+[Enter] on a Jupyter Notebook. SystemML seemed cool, but the only thing we ever did with it was multiply some matrices, and not even on a Spark cluster. I felt like I didn't learn that much because there was no point where I really had to engage my brain.

Some lessons are too hard to follow because of stong foreign accents and poor grammar, so the student ends up spending most of their effotrs on tranlation. Instructors mostly go through coding notebooks quite quickly. The oil forcasting module is an example of the defects of the course: Poor language, poor presentation, 'here', 'there', 'now' are used to reference the code with no pointers, slides are borrowed from online sources and insufficient. The proposed code is a tentative time series forecating toy example with catastrophic results, not at par with any time series standard.

The course is delivered at a very high level of abstraction. If you are a beginner, I wouldn't recommend this course as the explanations provided are quite vague and not so good in many instances. Justifications for the use of quite a couple of algorithms/values are not provided thus leaving the learner with a lot of "Why's"

One of the nice things about the course is that the instructor responds promptly to students' queries.

Again, the instructor speaks way too fast to explain anything. Even the subtitle cannot follow the instructor line by line. Frequent occurrence of inaudible words or sentence or wrong translations. When it comes to the code, never really understood what each line of codes is for...

Very good topic with not very good instructors. The exercises and programming assignments are too way easy and do not help in getting enough knowledge. All instructors are not native English speakers and have very strong accent and speak slowly.

It is difficult to fully understand the contents of the lesson, too many theories and not yet associated with practical problems. It's like studying in a university and for those who have more knowledge of math, not for everyone.

Explanations are a bit rush. Would not be easy to follow if I would not have deep previous understanding on the Deeep learning topics.

I felt the week 3 projects could have been given separate weeks to give better time for each project.

The progress through the material was good, but the delivery was quite boring.

The assignments were not good enough to test what was taught.

No pedagogy. No instruction, mostly copy and paste and guess.

Bem desatualizado.

Disappointing! The IBM name associated with the course raises a very high expectation. However, right from the start it is clear that it was done in a hurry with many typos, bad videos and recorded inside the car (amateur thing), the didactics of some instructors is terrible to be understood and translated into other languages. I believe the IBM name is associated with the courses to bring more credibility, but it's disappointing in the end. The material is very interesting, but the way it was delivered in this course leaves a lot to be desired.