It is very nice to have a very experienced deep learning practitioner showing you the "magic" of making DNN works. That is usually passed from Professor to graduate student, but is available here now.
While the information from this course was awesome I would've liked some hand on projects to get the information running. Nonetheless, the two simulation task were the best (more would've been neat!).
by Srikrishna R•
This course provides insights that you normally wouldn't get reading a book alone. While it does cover the core theories behind structuring of projects, what sets it apart is the truly practical tips and tricks that you could put to use in your project right away. The guidance is actionable and draws from practical experience of stalwarts rather than draw from theory alone. The test & exercise was quite innovative too as it puts you through a real world simulation to help you understand decision pathways you would take based on situational role play. Overall 5 stars!
by David T•
Having talked to someone who is actively working on Neural Network models, some of the insights I learned from the course looked to be helpful to them as well when we talked. I really appreciate the hands-on quizzes as well, as they gave me a chance to critically think through what I had just learned, and apply it to a real-world example. They especially helped when I got things wrong, because then I was able to rethink some assumptions I had made, and solidified my understanding of the material. I hope the next two courses are just as good as the last three!
by Don R•
This course ia about the practical application of Deep Learning techniques. Andrew Ng's other courses are very theoretical and prepare you with a very strong mathematical foundation for Machine Learning. This course provides practical advice and recommendations for teams building real-world applications of Deep Learning -- advice garnered over many years of work by Professor Ng and others, and, as far as I know, not collected into a single source anywhere else.
I have taken several of Professor Ng's courses. They are all excellent. This may be the best so far.
by Vishal R•
So far, this has been the most useful course out of this specialization! Sure, the others might offer more technical expertise, but this trains you things that cannot be taught in a class or a lecture. The application oriented case studies are extremely intriguing and challenging to a person whose knowledge might be completely theoretical. This course trains you to think in real life situations of applying a deep learning model, where to cut costs and effort, where to add more, how to optimize your model to surpass even the human level, and go further etc..
by kunal s•
It is one of the awesome courses everyone should join as by investing time for this course you may save your time in future when you are working on real world problems as Andrew has taught his experience where people makes mistakes and how to not repeat it and save your months of time,also he have taught in details about the datasets creation and there use.And also how u can use pre-trained model for other type of dataset. Join and it will make you more curious to dig dipper and also at same time making you better than some of real experts in the industry.
by Stephen O•
I took this course because I was leading a project, and so far so good, the learnings from this course have set me on the right path to troubleshooting the various problems we encountered, and as well conducting effective error analysis on the project. So far, based on Prof Ng's lessons, we have been able to improve our system by a certain percentage and come to a definitive conclusion on why some of our processes are not working as they should. This course is a must-watch for not just technical project leaders, but for all machine learning practitioners.
by Benjamin G•
This short course really fills in some gaps in terms of "tricks of the trade"; I think of useful information of this sort as the "force multiplier" whereby some small pieces of advice and insight from a practitioner goes a long way. I checked in a couple of machine leaning books and couldn't find equivalent advice. I particularly liked the point that was made about machine learning and certain ideas becoming obsolete (having previously done a PhD in machine learning) as I had that impression myself and was discussing it with a colleague this very week!
I like how it discusses everything on a strategic level. Very helpful when leading AI teams in the office. I wish there were a couple more case studies on different AI topics like natural language or signal processing or dialog systems. These are hot topics in the industry and academia and would be helpful to both professionals and students working on these problems to gain some insights to these problems as well. Thank you Andrea and Team! This is wonderful and would high recommend to L&D department to add this to our data science options
by Francois T•
0 math, yet I learned so much practical engineering advice, probably more than in all of the theory classes. I am very grateful. I love these kind of classes where we learn a lot from Andrew Ng's practical invaluable experience, and strangely, they end up more difficult than the pure math classes. To me, this kind of classes are higher quality training data for for the NN inside my skull than theory classes of something encapsulated in a library ;). It would take me decades of practice to reach these conclusions. Thank you for the wisdom!
by Jairo J P H•
El curso es muy bueno, particularmente estoy muy agradecido con COURSERA, por darme la oportunidad de hacer los cinco cursos de la Especialización en Deep Learning con ayuda economica y permitirme tener acceso a este tipo de capacitacion y certificacion. Muchas Gracias…!
The course is very good, particularly I am very grateful to COURSERA, for giving me the opportunity to do the five courses of the Deep Learning Specialization with financial aid and allowing me to have access to this type of training and certification. Thank you very much!
by Anders S•
Best applied course I have taken so far. Very practical, great to do before starting a project. I do have a suggestion for the specialisation in general. I have been working with deep learning, specifically image recognition and I have had a hard time figuring out what images I need to feed into my algorithms. Material about what type of data is needed to train algorithms correctly and overall requirements of this data would be great. I know this is done in some level, but not in a level of detail necessary for a project.
by Jon J•
Out of the deep learning specialisation courses so far for me, this has been one of the most valuable. Although iterating through solutions for deep learning problems is becoming faster, it's clear that huge amounts of time can be lost spending time chasing ghosts, as there are many areas in which empirical observation is the key to success. Some of the answers to questions posed in the quiz might seem counter-intuitive, but I believe the course sets up a solid framework with which to quantify solutions and approaches.
by J A•
Probably the best course to learn how to approach a Machine Learning project and deal with all the multiple challenges and issues which arise in real applications. Lots of years and experience of ML work distilled in a set of practical recommendations which can save one and entire teams months of work and computing expenses. The quizzes, based on simulated real cases, help mastering the recommendations. An ideal course for the more novice practitioners to catch up with the most expert ones in just a couple of weeks!
by Evandro R•
This is definitely a very important course for those who wish or already work with Deep Learning / Machine Learning. Andrew guide us into debugging and analyzing possible problems and decisions that we, as developers in Deep Learning might come to. The whole course is a very deep analysis on each step during a Deep Learning application, such as Error Analysis, End-to-End Deep Learning, Transfer Learning, Multi-Task Learning and others, it's a complete meal of information that we can absorb. Amazing teachings!
by Atul A•
Great course! This is the first course I've seen that gives a "big picture" overview on *how to approach* new machine learning / deep learning projects. It dives into how to structure the project, how to separate training / validation / test datasets, how to perform error analysis when your errors are high, how to trade-off bias/variance, and when and how to apply end-to-end deep learning. In short, this course is about finding the right trails, rather than going deep in the forest. Highly recommended! 👍
by Adam D•
Very good course! This course covers a lot of the nuance around actually running ML projects. It provides useful frameworks for thinking about how to proceed when you encounter different scenarios during the lifecyle of an ML project. Importantly, the course covers things like: how to think about human level performance on a task, doing detailed error analysis on your results so you can proceed in helpful directions, when to think about transfer, multi-task and end-to-end learning, and many more topics.
by Cédric v B•
This course contains some very essential information regarding the appliance of machine learning in a project. I think that it really discerns itself in this regard when compared to other courses. The lectures are very clear and I particularly enjoyed both exercises: the questions were very well chosen. Also, I quite like the 'Heroes' videos (also in the previous courses) as they also provide some very good information on the field of AI / ML in general as well as some practical tips on how to enter it.
by Phaneendra M•
One of the best courses I have ever gone through, the lessons were short and to the point thus allowing me to absorb the concepts even though they were bit outside my experience. Andrew generalized the topics so effectively that I could relate similar experience to understand the concepts. I love Andrew's simplistic, repetitive, regressive approach so if things aren't clear in the first go, you can trust him to reviw them at the right opportunity. I would love to learn more on this topic from Andrew!
by Artem D•
This is a very interesting course with very useful recommendations which could be also applied to ML projects. I highly recommend this material.
The only downside is that the course is structured as 1-1.5 hours of lectures and then practice quizzes (which are actually very interesting). And as for me, it becomes boring just listening without hands-on then, say, 15 minutes, despite the material itself is very interesting.
I hope that the next courses will have more practice.
All-in-all, a very good course!
by vineet s•
Very important course. Most of the stuff in this course is what is important for practitioners and is missing in other courses, I think most of the organizations and teams miss out on the strategy and devote far more time in wandering in wrong directions. It would be helpful if at certain points when referring to some concepts, a brief recap be given. There are too many concepts in other courses that at times you have strain yourself to recall. It is good mental exercise but it may help some folks.
by Martin K•
This course completely wrapping up the topics from course 1 and course 2 of the deep learning specialization while presenting up-to-date (and fun(!)) "real" word evidence cases. From all the courses in the specialization, I found this one particularly compelling in terms of easy-to-grasp and the best overview of ML projects. The assignments were outstanding, making you really the feel like you truly understand ML challenges, use cases and solutions to problems.
Totally recommend this course!
by Benny P•
This is a very good course on machine learning subjects that are rarely discussed elsewhere, namely managing machine learning project. And surprisingly, despite the easy feel of the subjects and their explanation in the video, the decision making that you have to take (and is tested in the quiz) in simulated project is hard. As project leader, given many choices of things to do, it's hard to decide what's the best thing to do, and this course shows, teaches, and trains you how to do that.
by Akash M•
This course is quite different from its counterparts. Firstly, this course doesn't teach you the hard and fast rules, that we are accustomed to in traditional computing. This course helps you develop intuitions about measuring the performance and efficiency of our Machine Learning System. This is going to be of extreme importance to all of us. The other courses can tell you how to design a system. This course will tell you how good/bad your system is, and how you can improve it further.
by Pantelis D•
Another excellent course by professor Andrew Ng, short, on point and clear videos.
There are no programming assignments in this course, the skills learned on how to debug / lead a machine learning project are being examined through 2 graded quizzes that simulate real world projects.
It is worth mentioning that some interviews with influential people on the field of DL are included and make the student fall in love with DL even more. Excited to see what's next in this specialization.
by Bruce W•
This was a good course, overall. It covered a lot of the decisions you need to make, when configuring and working to improve your neural network models.
There are not actual flight simulators. That is just how some of the learning exercises are described.
This course made me think about a lot of things--for example, is it better to simulate noise in "clean" data or to try to filter noise out of noisy data. Obviously, this course is just a stepping-off point for your own explorations.