No graded assignments anymore

Missing code evalution

Unfortunately, really disappointed with this course.

Having done the previous 2 courses int he specialisation I have come to realise that the courses are much more of a tutorial and could be seen as quick practice content before going for the TF Developer certificate or something. That in itself is fine, I feel there are other places to learn the maths/intuition behind DL (e.g. the Deep Learning Specialisation) but I feel especially in this 3rd course the content really doesn't justify paying for it.

For starters the explanations are very fast, hand wavy and don't go into any real depth other than just quickly explaining each line of the short notebooks (this can be useful). There is no discussion on how to improve the models or actually use this other than just pressing play in the notebook, the length of videos 17min per week is really not worth it, especially when better content can be found for free in Kaggle notebooks or on YouTube. There are also no graded exercises and after the first week they have given up on even providing suggested answers for the ungraded ones. The exercises don't test your TF understanding, just your basic Python loading of data and if you can copy from the example workbooks, they also have inconsistencies new untaught content and prone to errors that you haven't even been told could be an issue which means you just waste time being frustrated at not understanding what code you're even supposed to add rather than trying to understand the content.

I really like Coursera in general so this experience won't change that, but given that the instructor has free content on the TF website and youtube channel it seems like a waste to pay for this course IMO.

Hoping the 4th course will. be better

This specialization is false advertising. It does NOT prepare you for the Tensorflow certification exam. It’s especially disappointing after taking the fantastic specialization by Andrew Ng, and makes this specialization feel like a cheap cash grab by Coursera and DeepLearning.ai. This series of courses fails to prepare you for three reasons:

1 – The certification exam is done on whatever is the current version of Tensorflow (v2.6 as of writing). You can’t expect a specialization like this to update every minor release, but much of the coding is still on the v1.X version!

2 – The certification exam requires you to work in the PyCharm IDE. The IDE doesn’t even get a mention in this specialization and it is all done through Google Colab.

3 – The material is covered at a very superficial level. I was hoping to walk out of it feeling confident in using Tensorflow on novel problems, but I’ve barely learned anything about Tensorflow that I didn’t already get from Andrew Ng’s specialization. There’re a few minutes of lectures (some weeks less than 10 minutes). The programming assignments are either pathetically easy, or lack any guidance on what to do (seriously sometimes there’s no instructions at all, you have to guess what to do by the variable names), or both.

Save your time and money and go elsewhere to learn Tensorflow.

Be careful! This is not a four-week course, it is just a collection of short videos that you can watch during one morning. There are no practical exercises and the quizzes even repeat questions from one block to the next. The descriptive text is misleading, expressions like "you will build a" are used but you will not implement anything. Without a doubt, it is the worst course I have ever taken on Coursera.

Videos are too short. Unlike Andrew did, there was not enough talk on intuition and how to tune the hyperparameters. There are a lot of redundant questions in the quizzes, and not enough explanations on the notebooks. I would prefer graded exams, not ungraded ones with answers. I would prefer the coursera lab instead of the google collab platform as we cannot access again previous works.

Very little content. Extremely short videos, with some notebooks thrown in and show only the most basic applications... I would have expected in-depth explanation on Network-Architecutres, instead of having to memorize the name of some methods in keras (fit_on_text or fit_on_texts). Come on! This course seemed rushed, compared to the high standard of other deeplearning.ai courses.

The course is very basic and shallow. Neither gives us well working practical (or real life) examples nor details or understanding of the theory behind. There are no mandatory/graded programming exercises to practice, and the quizes are full with easy but meaningless questions. It was quite disappointing .

The teacher explains really great but the course is horrible, If I wanted to implemented myself I would not be able to it. Optional exercises? No explanation for most of the things. Not enough information to learn how to efficiently tweak the parameters

A lot of provided notebooks do not work - there are bugs, missing resources from internet (put out by instructor), mix of TF 1.x and 2.x which does not work. And also zero graded exercises. Even if it was free I'd say it's not worth your time.

Low effort compared to previous ones. No gradding assements all 4 weeks. No context on the pratice assements, links not working, videos to fill explaning things already explained, videos with low effort aswell...

Too shallow. The instructor spent too much time on some meaningless details and didn't explain some very important basic concepts nor dive any deeper. I hope this was taught by Andrew.

Really lacking cannot believe a such a poor course is assocaited with this program

The lack of graded programming assignments is a big bummer

No graded programming exercises in this course!!!

Not much lectures not much exercises.