It has a lot fo great content, but the way it's laid out confused me greatly.

It also felt very rushed at times, many parts weren't explained in detail, making many parts much less educational than I feel they should've been.

The best parts are where Lak is presenting, even in the lab solutions he went to more details than anyone else and made even those videos interesting after solving it myself.

The quality of the lesson material is great but the quantity is nowhere sufficient to get the hands-on experience

Labs were had less quality than in previous weeks.

My feeling is that this entire specialization is a glorified demonstration of what GCP can do with ML. The labs are not interactive and in some cases did not work. I don't feel that I have learned anything new. If I were to use GCP for ML purposes, I would need additional training to do it. I don't recommend this specialization

Labs vms are to slow. Speaker is difficult to understand. Mic varies and speech pattern is not clear. The presentations need some graphics rather than a guy talking. Sketch out the ideas on a white board rather than talking 5 minutes to a single slide.

The best course ever to give you the glimpse of the whole ML land - what is ML? Why ML? and most importantly How to do ML for real life business problems - from developing models to model evaluation, serving in production with all scaling tips and techniques with essence of Cloud. The best course for ML practitioners out there. Learnt a lot and what other place to learn from apart from the people who are actually doing it for the most popular products used across the web and mobile world.

This final course is also very good. Embedding is my favorite part and Lak is my favorite instructor.

Thanks Googlers! Looking forward to the next five courses.

Too shallow to truly be useful. I think if anything it gives you an idea of what's possible and roughly the areas you should explore and learn about but you won't learn too much following this through.

Good course, but I couldn't get over the Estimator API. IMHO it's too complicated compared to Keras and I just could not force myself to care about it.

Some lab contents are distracting from the core subject, deviate from the video contents.

I have learned a lot of things about Machine Learning from this course. Thanks, Google Cloud team form this awesome course. Actually there are many Machine Learning courses or articles out there. But nobody tells me what is the best way to do one thing, why I will use this step for certain things. If anyone tells but he could not show the proper visualization, it is the correct way to do so. Now I will be participating in real-life projects to apply what I learn. If I get stuck then I will use this course as my reference guide.

This course goes deeper into how to improve your Tensorflow ML models' performance without going deep into the models themselves. A good intro to learn how to systematically tune hyperparameters like the learning rate or batch size, and also how to start using regularization techniques and embeddings. Finally the course also opens the door to custom estimators, keeping you interested for the next courses.

Good class with a lot of information and new terminology thrown at you. Especially if you don't have a machine learning background. The labs are done in a way where you can get familiar with TensorFlow and Python programming, without having to know Python programming. Good intro if you want to be familiar with GCP ML. If you're looking to write a bunch of code this will probably not be enough for you.

This is one of the best courses in this specilization, as it goes deeper in the internals of ML and Neural Networks. Also, it presents two important features such as Embedding (creating sparse tensors from features) and Custom Estimators, which is the first step to get into advanced topics such as implementing new estimators from Research Papers.

I've really leveraged from this course a lot.

Thank you!

The only concern is that There are lots mention of Batch Normalization however nor instructor neither lab exercises point out how to do proper Batch Normalization using Tensorflow. It seem like they suggest manual scaling. Also best to include a section how to use TensorBoard to fine tune training properly. Not clear how to monitor and correct training parameters thru TensorBoard.

ML is a very rich topic with many paths to achieve a goal, that is important to differentiate because there is so much information people can get confused, this course focus helps to understand the topic. Labs range from basic to difficult and requires your full attention, there is still room to learn from other theorical sources.

Great course for understanding in and out of Machine learning model. I learn lots of cool thing in it. Most important I learn about Google Vizire: A great tool for hyper parameter tuning. Now I am starting "Advanced Machine Learning with TensorFlow on Google Cloud Platform Specialization".

Good class with a lot of information and new terminology thrown at you.

The labs are done in a way where you can get familiar with TensorFlow and Python programming, without having to know Python programming.

Awesome intro if you want to be familiar with GCP ML.

Great course! You get some basics on how to fine tune your model (and why those methods are effective). Nice introduction to NN and what I think was the most relevant: building estimators from scratch and how Keras can offer a simpler way to work.

I enjoy the lectures. The professor has a good speaking and teaching style which keeps me interested. Lots of concrete math examples which make it easier to understand. Very good slides which are well formulated and easy to understand

The best course of the specialization. Drills right into the heart of ML in terms of how to improve your models and also shines some light on how to create custom NN models which can be used with all of the prebuild functions of TF.

Some issues wit the Google Cloud Plateform (UI bug, instabilities). Aside that, the course is great and well built. Although I already knew ML theory, I learnt few tricks about implementation and running model at scale.

This course is of specialization is the most applicable in the case of my work. If customer estimator was explained in the earlier courses, many other experiments with our own custom estimators could have been tried.

Now that's what I am talking about. Great survey on nuts and bolts of ML practicalities. The how and why of model generation and manipulation. Very excellent. Super library of reference models and materials.

Amazing way to keep the audience interested. Throughout the specialization, I was always interested to learn what was about to come, next up. I completely recommend this specialization. It is a fun way to learn!!