I simply loved the course. I've been working with Machine Learning, but I didn't understand much about Deep Learning - this course helped me a lot to get started in this new research area.

This course give a good introduction toward machine learning and AI. someone who wants to pursue his/her career in ML and AI in future this course would definitely help him/her

I liked the pace and the tensor flow applications. This should be upgraded to TF 2.0 at some point. Also, I would've appreciated some GAN material.

I like this introductory course, very good one to start to learn machine learning. I will definitely continue studying and re-watch the videos.

Very instructional with lucid explanations, the hands-on practical or lab sessions helped me to actually practice what I have learnt.

Very Good course explaining the theoretical concepts related to deep learning . Thank you

A very fine tuned Course,used as a warm up course for deep learning,highly recommended

It was very good , understood all concepts and learned something new.

Labs are great hands-on training, but the lectures and lab texts don't sufficiently prepare the student for the assignments. Watching them and reading the text will not give the student the skills to solve the assignments, forcing the student to search online for a better tutorial. I recommend providing the complete code to create a MLP, CNN, SWEM, RNN, LSTM and GRU.

With a basic template created, the lab questions can then have the student change epochs, batch sizes, etc.

I am aware that some basic templates were provided, but providing a SWEM and having the student convert it to a RNN is a huge jump. I took detailed notes during lectures and read the lab notes, but ultimately had to find other resources to complete the assignments, because the answers were not provided by this course.

I hope this critique does not come across as a personal attack. I teach electrical engineering and understand how difficult it can be to fully explain complex topics. Thank you, all four of you, for creating this course!

I would like to thank Prof. Carin for a very lucid and intuitive explanation of the major concepts in Machine Learning covered in this class. This is the best explanation of the concepts of CNN and Reinforcement Learning that I have found so far !!!

I am also a little bit disappointed by the set of Programming Exercises at the end of some the lectures by other teachers. I think instead of giving students examples of programming using raw, low-level TensorFlow APIs because it overwhelms the main concepts. It is better to use high-level back end tool like Keras (NOT Slim !!!)

Course gives us the fundamental knowledges of Deep Learning (mainly) and Machine Learning, I see it very clearly and easy to understand, the Instructors are very dedicated, especially Larry . But the disadvantage of this course is that you maybe gain nothing about Pytorch, as all the lab/assignment are optional. Yeah, there are no practical lesson at all!

If you are the type of learning-eager, this course is also good for you. But if you want more pressure in learning that needs you to exert, I will not recommend it.

It is a very basic introductory course to important fields in machine learning. It tells important models like CNN and RNN and LSTM. but it does not go deeper into the technical levels of these models. Some parts about mathematics are not very satisfying. Also I feel like the course doesn't provide enough training for the coding work. Nonetheless, it is a good course to start with machine learning and the instructors repeat the concepts from the previous class, which helps me a lot in understanding the concepts.

Great course for beginners, did a lot to fill in the gaps in my knowledge. There could be a little more help with the actual coding parts of the project, the work done in ipython notebook is largely self-taught.

A good introduction to Machine learning, literally!

Wonderful experience and learning environment.

GOOD COURSE

The practicals are not at all aligned with their introduction. Specifically, in 2B you're asked to perform something that hasn't been introduced, and 3B could really use some hints. Also, you have no way to ensure that you actually complete the practicals as intended.

The codes can be explained in videos rather than giving them in texts in the open lab. This can make coding even more understandable and applicable.

i would prefer to be more practical, and the lectures to be step by step on how can practice machine learning in real dataset (problems). I see that generally coursera emphisises in the theory rather than practical.

When I came up to the assesment I quit, because I had to spend so much time to understand how the framework works.

I would say that is a course with traditional acadademic lectures, which is not my type.

Course is extremely long. Good luck getting through Week 1 and finishing it afterwards.

Nothing is well defined, it looks like the classes are taken from other programs and put together in this "introduction". For instance, you have a professor in week two whom was never seen before, saying that we addressed a never seen before problem "previously".

Not a comprehensive course.

W​onderful course! The low-level implementation code to understand what's really happening and the high-level abstractions to get rid of boilerplate code and focus on the execution is gold! It helps connect back with the theory/intuition beautifully. Only recommendation I'd have is to have GPU backed notebooks so that the runtimes are fast. Assign 4B timed out so many times, I lost count. Running the same code with minimum changes on Google Colab was *extremely* fast and only consumed a very modest amount of GPU resources.

I was an engineering major at Duke, but never took any sort of computer science/machine learning classes because I didn't have time. This class was super straight forward. Everything just made sense. I don't know how to say it other than that. It was great to see how much of the math and signal processing things I learned could be applied to something like machine learning. Before this class, I had no clue what machine learning was, and now I feel like I understand the main gist and the basis for all of the math behind it.

It's a very informative and well structured course for beginners. All instructors has made the entire course easy to understand with various real life examples and implimentation. I am very grateful to Duke University to come up with such an introductory course and I am thankful to all the professors of the University to make it easy for a beginner to understand and follow. And lastly I applaud for the coursera team for providing educational platform and resources for the learners.