This is a good beginner's course, but needs a lot of polish. The presenter is very knowledgeable, but his accent is severe, and on difficult words the transcript is entirely wrong, so there's no way of knowing what he's saying. Several of the reading assignments were mis-timed, some of the reading assignments either had dead links, or it was not apparent if there used to be a point to them but there isn't one now. The assignments were buggy -- I spent more time debugging errors in the Jupyter Notebooks that were baked in than on the actual assignments. The assignments themselves were overly easy, but the hassle of debugging made the assignments hard to get through -- the "TA"s didn't answer questions in the forums.

Good course content, but I frequently got lost by the organization of the datasets, files, etc. I learned to set up neural networks but I can't, for example, see how to run them on data on my own computer. Data is just magically present during the course.

Also, it would help to have the Laurence's notebooks available somewhere in the course summary. They are accessible in the unit AFTER he has talked through them in a video, but I always want to see them WHILE he's talking through them (not just the image in the video, the actual notebook), not afterwards.

Some course material is mixed up (e.g. MNIST and Fashion MNIST datsets and examples are 'convoluted';-). Also, the performance of the resulting models is overstated. I don't think it is very impressive to make models that still make mistakes such as qualifying a horse as a human (or worse: an attractive woman as a horse). The idea from the media is, that computers/algorithms beat humans in image recognition easily (e.g. recognizing diseases in medical images), but this is not evidenced by the contents of this course.

The course is fine, but quite basic. I didn't like the fact that there was no way to submit any code homework (as in the other deeplearning.ai specialisation). Just reading some code and experimenting with it is not a good way to learn in my opinion. A suggestion: show a picture depicting the NN that is being built in the code so *this* code implements *this* CNN (or whatever) with depictions of the NN structures as in the other specialisation.

The assignments were a bit outdated and a bit difficult to understand due to the automatic checking ... you need to improve the assignments and tests to make them more user friendly and to evaluate more effectively the topic what the student has learnt(it was becoming difficult to pass not due to the code learnt but due to the automation , memory issues, and the evaluation code. But the Discussion forum helped a lot though!

I consider programming assigments could be better. In comparison to the deep learning specialization assigments, the ones we had in these course are really poor in terms of information and clarity on what we have to do. Specially, it is incredible that there are some unfunctional code on the notebooks such as the javascript part to save the notebook, it only caused error for many students as seen in the forum

This is a great introduction to Keras, and I learned about some unknown features. Unfortunately, I had thought it would be more focused on Tensorflow, since it's in the title of the course. I had decided to take this course midway through the Deep Learning specialization. I was hoping to gain more practice with easier Tensorflow examples, but the course didn't cover any core Tensorflow.

Very approachable course, probably a little too much. Assignments can be done by simply copy-pasting notebooks from the videos without any modification in the model. The generated images are also not that varied, and give skewed image of what humans should look like.

Anyhow, the videos are good, as well as the annotated notebooks, to familiarize ourselves with CNN and the Keras API.

A good quick walkthrough of how TensorFlow works, but not very in-depth. Definately worth taking, but not worth the $60 for the certificate in my opinion. My 3 star rating was for two reasons. 1. Price was too high for what was taught. 2. I feel like he could have gone a little more in-depth on how some of the functions were working, maybe had more complicated exercises.

The course can be done in 1 week or less.

Graded exercises end up being a variation of the already available week's notebook + a bit of code that is also repeated among every exercise. Not much to learn there. First exercise was a bit confusing in its wording also.

I think we should demand more from these courses if we want MOOCs to be taken seriously.

A very superficial course for anyone with even a little experience in TensorFlow. Can be completed in less than 6 hours. Overall it was a pretty good refresher for me, but I was inconvenienced by the poor design of the assignments. I completed the deeplearning.ai specialization on deep learning previously and it had terrific assignments.

I liked the instructor, but the way the course was put together is little awkward to me. Especially reading tasks, they give you a few sentences and ask you to mark as complete, and this is the end of the task. What's point? Also, i think the coding tasks should also be self-sustaining. You can still learn through this course, though.

The course title should indicate that it introduces Keras API NOT TensorFlow. The course seems to be focused on students who are new to neural networks. However, it does not explain any concept with detail. On the other hand, students like I, who have already completed Deep Learning Specialization would find this course too basic.

This is my specific review. I have done a lot of deep learning before and doing tensorflow , thought of getting rigorous exercises but there are only 2 examples. May be this is not the right course if you want to become expert in tensorflow, but it definitely has some organized information though to start off.

The course a bit weird in the sense that if you have some knowledge in deep learning (i.e. you have passed the deeplearning.ai courses) you would learn almost nothing. On the other hand, if you did not take any other courses or do not have any knowledge you wouldn't understand whats going on.

I think there is not enough of background explanation. I did the exercises and they were easy, but I still don't understand the image recognition very well neither TensorFlow. But I did get some starting point to continue learning on my own. This was my first interaction with TensorFlow.

This course is really hands-on and quite easy. You'll be using Keras during the course and it is assumed you had completed the Deep Learning specialization or another deep learning course previously. You'll be able to finish it in a couple of hours. So grab a beer and enjoy the light tour.

The course speed is relative high for an introduction. I, personally, have covered most of the contents using Keras's master book "Deep Learning with Python". I would have expected more insights into TensorFlow and not out-of-the-shelves code using Keras sequential models to play around.

Course materials are great, but there is a minor mismatch between Jupyter and Collab in versions of code and TensorFlow, as well as some comments in code, do not match the actual code below in constants. So it would be good somebody could check it through to make sure all is right.

A bit bored of only doing image recognition in every TensorFlow tutorial under the sun, including this course, as if there are no other problems to solve with it. The title is so broad that frankly I was expecting a bit more for the money than just doing image recognition.

Instructor does sound job of explaining how to use TensorFlow for deep learning. Compared to Andrew's courses, much less content in the videos and exercises. I feel that much more thought and effort could have gone into this course, just look at how verbose the name is.

it would be ok if this course were free but it's not. there are technical mistakes in the multiple choice quizzes and the mandatory programming tests. some important topics are left unexplained while some are repeated unnecessarily. check the forums for more feedback.

Though it does give you a good introduction how to make neural networks, i felt that this course doesn't explain the Keras API library, which makes it difficult to remember and understand how things work.

I hope further courses in the specialization take care of that

I feel there are gaps in this course. But it was still worthwhile going through the material, to reinforce certain concepts. I had the impression that the course was initially classified for beginners, and that later on that classification was modified to advanced.

Rather basic, but the learner can strengthen it by broadening his reading. What I missed most is a conceptual introduction of what Tensorflow is and does.

I did not take the Deep Learning specialization before this one. Maybe that would have made a difference.