Learner Reviews & Feedback for Introduction to TensorFlow by Google Cloud

nice!

good

Pace is too fast. It seems easy in the beggining, but when you get to the Challenge exercises, you see that you are lacking a lot of basic knowledge. It took me 3 hours to understand how to do a loop, using the low level API! I couldn't understand exactly the challenge of week 2 and post a question in the forum. No mentors answered it.

I didn't like week 3 class. The guy showed the source code and just made a quick comment about it, not explaining what was going on. I had to pause the video to check what was there, but I'd like actual detail explanation on what is going on there.

So, what I think is that this course needs more detailment and more attention to the questions in the forum.

The first 2 courses in this specialization had too little substance. Here the first two weeks are very good. Unfortunately, the final week terrible. It introduces lots of interesting concepts but does not explain them adequately. You will see lots of cryptic code (getting obscure ids, running bash scripts from jupyter) without much explanation. The whole thing feels unfinished. When google is not too busy being subservient to China, it could maybe spare some resources fixing this for paying customers?

It's an OK course, the examples are useful, but the labs often miss the balance of providing enough direction to guide the student, without telling the student what to do.

Content varies in quality and feels like it's patched together from a set of other offerings, meaning that there are occasional gaps (especially in the lectures) in the coverage.

As a course to complement other sources on ML and Tensorflow, it's OK, but I wouldn't take this course as your sole intro to the subjects covered.

I took this course separate from other courses in the series. So my review could be different if I have had all courses. Although the material covered are generally useful, I believe the course content does not help much with learning TF and code using it. Overall, the course flows well. But the third week is radically different from the first two weeks and difficult to follow.

Pros:

The first course in the specialisation that actually prompts you to write code for yourself.

Good concise introduction to TensorFlow, well written and presented content in the videos

Cons:

Knowledge of ADVANCED Bash and ADVANCED SQL required for Labs otherwise the whole thing goes over your head and you just end up copying the solution from the GitHub repository.

It was a bit challenging to write some of the Tensorflow code myself, I felt unprepared when attempting to do that although I was able to understand what was going on. This made the challenge exercises pretty intense. Also, I had some weird bugs in my Tensorflow assignments that didn't allow me to complete the last two labs really.

The course lacks detailed description of how to deploy your model on Cloud. The 3rd week of the course is useless. The course creators should've either concentrated on just TensorFlow for Machine Learning or should've provided detailed lectures on deployment of the Model on cloud.

Material was too high level. Talked more about specific examples than going through a broader understanding of the tools. It would have been helpful to start with an example, talk more about the tools themselves in detail, and conclude with a few more examples.

It was a little bit tough for beginners, but still a great course to learn. You must learn a lot of things on your own, after all, these courses are only meant for showing the right direction, which they successfully did. Thank you.

Introduction was quite good

Learnt a lot from that

but after week 1 it was too boring we were taught something else and labs we based on some other complex things. Make sure to improve the quality.

Good content and excellent theoretical explanation. lab materials are not updated. Running into too many Attribute errors and tf doesn't contain ** module.

Lab directory has to be updated.

The content is great, and the instructors do a good job. However, I could not connect to qwiklabs, so it would be beneficial if Google would create labs within Coursera's framework.

I learned a lot as it is more interactive. But Several problems with the lab and incompatibility with the instructor labs. I think the videos are older than the labs.

The course focuses a lot on boilerplate code rather than concepts, the labs are too broad and don't focus on learning specific topics about TensorFlow.

Conceptually, it was a good introduction to tensorflow. Labs were not introductory labs, there was heavy coding, and a few errors in the instructions.

Los laboratorios eran demasiado restrictivos, para poder crear un "bucket" requería de un permiso adicional y la práctica la tuve que abandonar.

The first week is okay. Mostly a waste of time if you want to learn tensorflow however--it's really just a huge ad for Google Cloud Services.

Good basic intro. I now understand what tensor flow is! But the second half tails off and is a bit disappointing. Some labs need updationg.

There is little to no visual illustration on how data structures get sequentially transformed with various estimator API functions.

Started very well but got quite confusing towards the end. I would not be able to produce the code in the last lab myself.

I think that the instructors were good but content was not sufficient. Also they culd have added more on tensorflow2.0

Good intro to TF, but not updated for 2.0. Also felt at times more like a lesson in Cloud Platform and Jupyter.

The course is just gives an insight. It would have been more beneficial if there were some graded assignments.