The videos are ok, the "Labs" are really bad. You just follow instructions with code to copy into the notebook. Of course, you can play a bit with the code, but you don't really learn how to do it yourself when the correct answer is already filled in. Would not recommend.

Great course. Gradient descent and loss function concepts were explained well.

Kind of tough for a beginner course. Lots of vocabulary to understand and some of the quiz questions were worded in a way that made it so there were more than one good answer to the question.

Amazing course. For a beginner like me, it was a shot in the arm. Excellent presentation very lively and engaging. Hope to see the instructor soon in a another course. Thanks so much. I learned a lot.

Good but does require quite a lot of knowledge of Python and coding, which was not exactly expected from the start (though it's a good balance between complexitx and ease-to-pass for the course even without this)

The course is ok. Several complicated concepts are expected to be known, other very easy ones are explained in detail. However in some phases too high level, I am definitely missing some course resources to work with.

Was hoping for more hands-on experience.

Someone really needs to proofread the quiz questions, and fix the links to datasets in the lab. Apparently some UI was decommissioned as of October 1st and the datasets don't open in the browser at all. Also, providing only 45 minutes to do labs where you need so much more time isn't helpful. Someone should fix the python code, too. It looks like it's using R conventions which is very confusing.

Although there are engaging lectures, they skip over all interesting technical detail to focus on philosophy (and of course lauding Google). Philosophy of data analysis is all good and well, but there is no technical instruction to go with it. Quizzes and labs stress technical details for which there is no instruction. Readings are a hodge-podge of unlabelled links with no prioritization or curation.

Labs take 15 mintues to start up, and then have 1-2 hour timeouts, at the end of which you lose all work you have done. There is little to no instruction in the labs: the only way to do them is either already understand what their topics (making them pointeless) or to quickly google up answers to technical questions, and blindly run functions (which is also pointless).

I have no idea how this course sequence was ever rated as an interesting one for qualification.

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.

looks more like a promotional course from google instead of an acutal learning experience.

Also the labs have no data on the code used, it is assumed the learners are well acquianted with the technology used that is specific to google.

i cant un enroll

Great lectures and labs thanks. The first lecture block made a lot of great connections between topics and methods past and present yet get the most out of it, one ideally has recently reviewed the theory behind the tradition tools. Otherwise the first block is a bit of a drink from the firehose although one can still pick up the gist message but may not get some of the other enriching points. In any case. Great work and thanks

Today is my day Learning GCP AI platform have a fun time discovering data pre-processing with Big Queries, deploy TensorFlow notebook and play with Benchmark model. The fun time is having a chance to take a look at the Google Cloud AI platform and have a fun time with it😊 ! Thank you Google and Coursera give us the scholarship to read and have a fun time with These 🧪 🧫 labs☺️! You light 💡 my day🍀! We love you ~~✧٩(ˊωˋ*)و✧

Learned about the concepts of ML Molde optimization, model performance matrix, model generalization and sampling. The general concepts are well explained so that I can better understand how each of them is done. Tensorflow playground is a very interesting part of exploring tuning hyperparameters, and the ML history go through is also another highlight. The lab at the end is a must-try learning experience.

This course builds on the previous one. Although use is still made of Google cloud, the course becomes more interesting, since the teachers provide their practical insight in the preparation of data for machine learning without focusing too much on Google. The history of machine learning is very interesting and the labs very useful in understanding the main pitfalls associated with the preprocessing step.

Great course, thank you. However, there are some things that should be checked again. For example, there are some quizzes that have questions which answers are not in the videos. If those answers are in the readings, a better course organization should be useful to know this. Also, there are some parts in the labs that require also more precision in order to know what specifically is asked.

An excellent introduction to Machine Learning, I appreciated the explanations around the importance of having proper training, validation and testing set to build robust models, I loved the introduction to Big Query and the value of cleaning the datasets, plus all the explanations around Classification Models,Regression Models and Gradient Descent.

Thanks

Thanks to team of Google Cloud Platform for giving such dramatic and interesting course for me to acquire critically fundamental knowledge of conducting ML! It leads to all the places I've never thought of, and now I'm prepared to accept the challenge to bring ML into solving real-life problems, in the hope of making the world more sense!

It's a good course on the principles required for the ML model building. Repeatability is important in machine learning and it can be achieved very well using BigQuery.After all, it's a course offered by Google which has been outstanding in the deployment of ML models in every aspect of its product. Go for it it's worth.

Fluent flow of introduction using examples. Gives an overview of the ML process concepts and tips. Detailed concepts are only mentioned quickly, so to fully benefit from the course would probably require some prior experience in ML. Valuable insights and summary from experienced ML engineer are provided in this course.

I love the course introducing core concepts and practices in machine learning today as well as some historical development. This course feels more rigorous because some core mathematical foundations are introduced. I kind of hope there could be more theoretical explanation in more depths with some references attached.

Excellent course to understand how machine learning is done , classification, regression, RMSE, cross entropy, gradient descent, loss functions, performance scores, splitting data into training, validation and test data sets consistently/repeatedly using mod and hash functions, exploring, cleaning data.

I loved TensorFlow Playgrounds. It made so many concepts visible. I have more intuition into how the number of layers, input features and number of neutrons affect what the model can learn. This is not my first Machine Learning course, and it is helping me fill out many gaps I have in my understanding.

Thank you for such great knowledge sharing. I have really enjoyed the course and have learned a lot from it. The way the speakers explain each and every tiny detail is exceptional. This course make me a step closer to my goals and will help me in my career building as a Machine Learning Engineer.

This course is insightful. I'm new to SQL, so I couldn't understand what was going on in the lab. I still find it enjoyable, and I think I've learned a lot. Maybe I'm not able to know how to split data as good as the instructors, but I'll use the resources on GitHub to keep improving.