Chevron Left
Вернуться к Машинное обучение

Отзывы учащихся о курсе Машинное обучение от партнера Стэнфордский университет

Оценки: 166,828
Рецензии: 42,707

О курсе

Machine learning is the science of getting computers to act without being explicitly programmed. In the past decade, machine learning has given us self-driving cars, practical speech recognition, effective web search, and a vastly improved understanding of the human genome. Machine learning is so pervasive today that you probably use it dozens of times a day without knowing it. Many researchers also think it is the best way to make progress towards human-level AI. In this class, you will learn about the most effective machine learning techniques, and gain practice implementing them and getting them to work for yourself. More importantly, you'll learn about not only the theoretical underpinnings of learning, but also gain the practical know-how needed to quickly and powerfully apply these techniques to new problems. Finally, you'll learn about some of Silicon Valley's best practices in innovation as it pertains to machine learning and AI. This course provides a broad introduction to machine learning, datamining, and statistical pattern recognition. Topics include: (i) Supervised learning (parametric/non-parametric algorithms, support vector machines, kernels, neural networks). (ii) Unsupervised learning (clustering, dimensionality reduction, recommender systems, deep learning). (iii) Best practices in machine learning (bias/variance theory; innovation process in machine learning and AI). The course will also draw from numerous case studies and applications, so that you'll also learn how to apply learning algorithms to building smart robots (perception, control), text understanding (web search, anti-spam), computer vision, medical informatics, audio, database mining, and other areas....

Лучшие рецензии

2 апр. 2018 г.

Very nice course,. Give a fundamental knowledge of machine learning in a clear, logic and easy-to-understand way. Suitable for those who has relatively weak background of math and statistics to learn.

16 мар. 2021 г.

I want to thank you very much for such a great course in any aspect especially from professor Ng . I just want to suggest that it would be great if there was a final project for the end of the course.

Фильтр по:

601–625 из 10,000 отзывов о курсе Машинное обучение

автор: Mouhamadou M S

3 авг. 2018 г.

Going through the whole process, I really think that this course is a strong introduction (and sometimes more than just an introduction) to the machine learning field. Andrew gived to us many advices about the applicabilty of the learned concepts that could help anyone to get more confident while having to conceptualize the ML problems and to implement the adequate machine learning systems. However, I would have appreciated that some of the scripts that were already implemented would have been part of the exercises to help us enlarging the technical skills gained using Octave/Matlab. Specifically, some of the plotting scripts or the script to implement the pre-processing of the e-mail text (spam classifier exercise) could have been opened to some stepwise implement to participants. Nevertheless, my general impression of the course is extremely positive.

автор: Sebastian

17 окт. 2020 г.

Hello, I just want to say that this course was extremely interesting. I am a phd student in physics and tried to find a way to learn machine learning by myself. Luckely, one of my colleagues recommended this course and so I started and enjoyed it very much. I think that I learned quite a lot and are now eager to apply this knowledge to my own projects and finally also help other people. For me this course was the perfect starting point and I am sure that I will remember it for a very long time. I also would like to thank you that you provided all this material for free. Although I paid for the certificate I know a lot of student who might not have the ability to pay money for special books or so. I hope that I can also recommend this course to a lot of future students and wish you all the best. Best regards and thanks for the beautiful course! Sebastian

автор: Pavel T

23 авг. 2017 г.

I wanted to say how Iam grateful for this oppurtunity to take this outstanding course. Honestly, that was best class I've been through for all my life. Proffesor Andrew Ng has really wonderfull style of teaching, its a big honour for me being your student. I wish someday I will be able be your student in real life. I understood every aspect of this class, all explanations were clear and repeating. It made me clear up and feel all maths and technicalities. Programming exercises were great, I saw how algorithms are working in my PC downrightby my commands. Andrew Ng is a machine learning developer, so his stories of live examples were awesome.

Thanks to all mentors. Thank you for responses and all your support. Special thanks to Tom Mosher, he resides this course and his support means a lot for all community, Tom is great in making tutorials.

автор: Vladimir B

11 февр. 2018 г.

I found this course has a good pace, I feel like I've learnt a lot but at the same time nothing was rushed and I never felt like I was struggling to understand. I think many of the examples used were well considered and give you confidence that you could actually create your own machine learning algorithm for your own application. My only criticism would be with some of the programming assignments, where I would sometimes spend more time trying to understand the existing code (so that mine would be using the right matrix indices for example) than actually doing the maths and learning about the intended subject. I don't know how you could get around this problem without requiring the student to write the whole program themselves though, which I certainly would not prefer! Maybe this is an inevitable consequence.

Thanks Andrew I really enjoyed your course!

автор: Skltfz S

24 нояб. 2016 г.

i learnt something, but i can never able to use it. i have no idea, anyway, ML is just still having the real application for the command development field in the business now. for example like me, working in software development firm and medicare company before, none of them really applied the machine learning theory. although, they have certain value if applied, for example i saw many of them trying to convert their business to big data business, however they never use the big data correctly, assuming applied a machine learning in order to predict the potential customer response of a single product, it looks fun. anyway, fact is fact, currently the business is still a relatively low technical level business. everything is report, linear, and simple ( i will not say unfortunately, at least i have time to write this comment when i am actually working)

автор: David K

5 февр. 2020 г.

I really enjoyed this course, Andrews bottom up style gives you enough detail to understand the key concepts in machine learning and lay the foundations for further training. Andrew shields you from the hard maths but creates an understanding of the intuition that makes you feel you know the maths.

I would have preferred python for the code but in my view the value was in the wider machine learning training as opposed to coding. That said it was very satisfying to code an algorithm and see it work, this level of detail helps to understand libraries like scikit-learn.

My motivation for this course was to understand what’s under the hood of machine learning and this course delivered that. Overall, this is a great course and I learned an enormous amount of information, its the perfect compliant to the more practical and hands-on courses I am taking.

автор: Tomáš D

25 апр. 2019 г.

This course provides an excellent introduction into machine learning and is a great resource for anyone looking to get into AI/machine learning/peep learning. Most of the methods and concepts are useful even in today's deep learning era. I would especially like to point out prof. Ngs excellent advice on applying the taught algorithms and general best practices.

The assigments are in Matlab/Octave of which I was sceptical at first, but soon realized that the in-built matrix operations of the language make it ideal for implementing learning algorithms from scratch - so that you understand them in-depth. There is also an Octave tutorial in the course which I found sufficient for the assigments, having no previous experience with the language.

Overall, this is a high-quality course that I would recommend to anyone wanting to learn about machine learning!

автор: Syed M I

9 апр. 2019 г.

I was wondering why we are being forced to use Matlab/Octave in a time when everyone that i know uses Python. I realized it's wisdom in the first week itself.

Mr Andrew has designed the course very cleverly, he is aware to teach things in the least time, in the easiest way possible. He boils down his theory to few lines of formula which can be coded in a few minutes. Throughout the lecture he keeps on giving hints on how to structure the code for that algorithm.

In the programming assignments also he makes sure that you only get to apply your brain where it is needed and he has himself pre-written the labour intensive part of the code. However we can easily analyse the whole functioning of the code by merely glancing at the portions which we have not written.

I am about to finish week 4. He presents this subject like it's a standard 10 school subject.

автор: Charles B

8 июня 2018 г.

Andrew Ng presents the major machine learning algorithms, explaining them from the ground up. The level of detail in this courses was great from my point of view: it skips some of the calculus (e.g. doesn't show working for most derivates), but does explain the linear algebra involved. This is important because the parallel processing needed to implement machine learning efficiently relies on use of matrices and vectors - as done in the coding assignments in this course, which were a lot of fun to solve.

The assignments are in Matlab, which isn't a mainstream programming language; but I found it easy to learn and it does lend itself well to applying vectorisation techniques. Overall, I found this course really engaging and I now understand details of algorithms like linear/logistic regression and neural networks that I'd previously taken for granted.

автор: Ryan M

7 февр. 2016 г.

This is a truly superb class. Professor Andrew Ng's lectures are clear, well-organized, and exceptionally informative. He's more than a brilliant researcher: he does a terrific job of presenting complex machine learning concepts in simple terms that a very easy for anyone to understand as well. I would add that the programming assignments are also very useful because they focus on core concepts and really help to reinforce the lectures. All around this is easily one of the best and most valuable courses I have ever taken, and I would be very quickly to recommend this course to any friend without reservation. Whether you're actively working with machine learning systems as I am or hoping to get into the field or merely curious, Andrew Ng's machine learning class is a very valuable class that covers the essentials and covers them very thoroughly.

автор: Jordan S

3 мая 2020 г.

This is a truly great course on a fascinating and important topic by one of the world's leading experts. Andrew Ng chooses the most important topics in a field that is rapidly evolving. He goes into depth on various supervised and unsupervised machine learning techniques. He discusses the mathematics of the algorithms, how to implement it in code and examples of real world applications.

The coding exercises were surprisingly challenging. The amount of code students need to write is quite small, but the matrix operations can be tricky to get right. My only real issue is this class is that the coding exercises are in a language called Octave. Andrew Ng defends this decision by saying that is is well suited for introductory level topics, but I still think it would be better to teach the course using a more commonly used language such as Python.

автор: Yan L

13 мая 2017 г.

There are tons of ML classes on internet (even for free), but this one is the A)most comprehensive topics coverage, and B) easiest to understand.

I am not saying your will become an expert of ML (obviously) but you would get a full picture of what ML is doing and how will they improve your daily life and work. You still need a lot of math/programming experience to build your own ML system or application software. Having said that, by taking this class, you know where you should start. The 2nd comment is of course a compliment, and requires years of hard work and practical experience to explain such a esoteric topic to everyday people like you and me through simple word and a well-structured agenda.

If you have zero knowledge in ML/programming/calculus/linear algebra/statistic like me, please take it and am sure you will get more than you expected.

автор: Jathavan S

7 янв. 2017 г.

I think this course is the best entry into the concepts of ML you can find. Andrew Ng is a wonderful, passionate teacher and explains most concepts in a way that is easy to grasp. I liked the emphasis on supervised learning and taking time to explain the very foundation with linear regression, logistic regression in great detail. When you look around the internet everyone is talking about cNN, Deep Learning, GANs, ... but the truth is - you first need to get your foundation right before moving on the more advanced topics. ML is actually a lot of mathematics, stochastic - so getting initial knowledge about these topics is important. Some things in the course where not that well explained, for example Back Propagation. In general I can recommend this course to anyone who wants to START with Machine Learning and needs some orientation on the subject.

автор: Alexandre S

17 янв. 2021 г.

Overall, the course appears dated but is very interesting, very informative, very well taught, and very relevant. The course is a bit old now (2011?), which is apparent in the relatively low quality of the video and audio. But since it deals with the basics of machine learning, it is still 100% relevant to 2021, and the audio/video quality is good enough for purpose. There are some minor mistakes here and then that could be corrected, but it is understood that Prof Ng has moved to other ventures and so won't be recording the material again. Mentors are quick to answer and help students. Quizzes and programming exercises are of suitable difficulty. I did the entire course at 1.5x playback speed and could finish it in about a month, partly over holidays. Only disappointment is that there was nothing on decision trees (Random Forest, Boosted trees).

автор: Jimmy G

10 дек. 2016 г.

This course was amazing in many different ways:

1. I really learned a lot about how machine learning works (at least for the algorithms covered in the course). And I'm really keen to continue learning.

2. Learning tools and methods: anyone could just read about ML and algorithms, but the way the course is focusing on exercises really helps assimilate the content

3. A lot of effort goes into providing support to the students. I'm very thankful to the staff who are VERY responsive on the forums. They also provide a lot of test cases and other help to better understand how to complete the exercices

I recently left my job to focus on personal development and am doing a lot of online courses right now. This course by far the best one I did so far. I'm really glad I took the time to do it. And looking forward to learning more on the topic

Thanks again!

автор: Ankur S

16 июля 2017 г.

This is a fantastic course for beginners. The only pre-requisite you need are fundamentals of high school math (Matrices, Vectors & their operations) and some basic computer programming knowledge (any language is fine as long as you can understand variables, arrays, for-loops & functions). Prof. Andrew Ng's teaches basic & advanced concepts in a manner that is easy to follow. What helped me the most were the programming exercises at the end of each chapter to help understand the different algorithms and the parameters that define them, better.

A few recommendations for this course.

It would be great to include a couple of more programming exercises, especially for un-supervised learning. If not in the main chapter, then maybe in the addendum.

It would also be great to have the concepts tested with more questions at the end of each chapter.

автор: Rubén C

31 мар. 2020 г.

This course is a great introduction to Machine Learning. It guides you through the most relevant machine learning algorithms and techniques and gives you insight into the mathematical essence of each algorithm, until the point that you will be able to program them (a basic version) yourself in the Matlab programming language.

It also teaches you how to approach large scale machine learning systems (at a general level), and gives you practical tips for evaluating your algorithms, how to work with real-time incoming data, among other useful concepts.

I had a great time. Andrew, the professor, is charming and you feel that he truly enjoys teaching. He is so good explaining the concepts in a simple way. In just a few minutes he is able to teach you complex concepts. He gives you both the intuitive understanding and the mathematicaly formal one.

автор: Adrian L

4 сент. 2017 г.

Great introduction to machine learning. The videos were very good at breaking down the different concepts and algorithms. It was very helpful to have summary notes available. The quizzes were useful to consolidate knowledge. The programming assignments were at an appropriate level of difficulty, for the most part, where they required some thought but were doable within a reasonable amount of time for a beginner.

The one thing that annoyed me is some of the videos were somewhat sloppy in terms of editing. There were parts where some of his narration was re-recorded but not spliced together properly, such that it was repeated. There were also a fair number of errors in the slides, but these were mostly corrected in the errata. Overall, not a big deal, but it seems like it shouldn't be too much work to just splice the clips together properly.

автор: Ignacio F M

16 апр. 2020 г.

It is a great course in order to begin with Machine Learning. Covers every basic aspect of this field, even every some more advanced topics, the explanations are very good and the practical exercises are interesting. The course is well suited even if you do not have a good level of programming or Mathematics.

Though, some of the aspects that I think could be improved are that the questionnaires are maybe too easy and short, and that in at least one of the practical exercises the student should work on how to build the script to put all the different functions to work togheter.

And if you already have knowledge on Machine Learning and your level of mathematics is higher, then the course is still good since it will cover all the basics and fill some gap that you could have, even though there might be more appropiate courses on this webpage.

автор: Saravanan T S

18 мар. 2017 г.

This course has been highly delightful to learn through the concepts of machine learning. For someone like me having 17years programming experience, and with some hands on of neural networks from colelge days, and analytic tools for a few applications in R; it is a great refresher on the fundamentals and great breadth of practical technology elements that are most useful in machine learning applications.

Andrew's teaching method is great. His clear and simplistic delivery style of complex concepts with apt examples ensures the student grasps the essence with ease and works navigates his way through complex algorithms with confidence. I would love sitting through his lectures in the future too!

Congratulations to the entire team that put together this course and making this a great service that is available for anyone wanting to learn ML!

автор: Justin Y

9 февр. 2019 г.

This course is wonderful and charming. By taking this course, I got the basic knowledge of machine learning and artificial intelligence. Concretly,I knew what is supervised learning and unsupervised learning and I also learned how to operate that by myself through a fresh language called Octave,which I never used before. Besides,I also want to express my thanks to Prof.Andrew Ng for his kindness teaching and shareing his knowledge in this field. Not only his broad knowledge but also his skillful teaching method that impressed me a lot. He could always taught us many difficult and obscure conception by taking some simple and clear examples,which can let us understand easily.

I will continuously take a course by Andrew Ng called Nueral Networks and Deep Learning, and I will dedicate myself into this course and I hope I could learn a lot.

автор: Francesco P

14 нояб. 2020 г.

Dr. Andrew Ng has the unique ability to explain complex and articulated concepts with incredible ease and effectiveness. Throughout the entire course there's never been a moment where I felt lost because each topic was explained very well and even though several years have passed since this course was released, it remains maybe the best one to get both all the theoretical foundations of Machine Learning and getting the hands dirty by implementing the algorithms in a low-level fashion using the Octave language. It certainly is difficult but it helps you understand all the little things that otherwise would've been difficult to understand by only watching videos about theory.

It's been an amazing journey, thank you to both Dr. Andrew Ng and the amazing community of learnerns that made the learning process a little bit less hard.


автор: Angadbir S

19 апр. 2020 г.

The course is an excellent introduction to the Machine Learning techniques. While the field is evolving by the day, having the experience to code the basic and powerful algorithms by hand provides a fair confidence and intuition of the inner workings of these cool-sounding techniques. The subtly introduced idea of Matrices as efficient computation model was very interesting having read it back last time in high school math without knowing what their real life application was. Immense gratitude is in order for Andrew Ng (and probably team) for having created the course that helps one focus on coding the actual algorithm rather than the data preparation and other data flow problems (which I understand is more than half the time spent for a practitioner). This is a must-do course for people thinking of starting to learn about this field.

автор: Rohit K

17 июня 2019 г.

This is the course for which I have joined Coursera initially. Professor is very elaborate in explaining anything. I am a student from core Mathematics and Statistics background. But I feel that the course is designed in such a away that any students without having a high school level knowledge of basic mathematics and programming can grasp the ideas discussed here. But I would recommend this course as an introductory course to those who are rather interested in the core mathematics behind Machine Learning. For me, It was a good course to get a overview of Machine Learning as I didn't have a proper course in ML earlier. The positive thing is, a positive interest have been built regarding ML as I was learning from Professor Andrew. I will definitely go further to learn deeper in mathematics of ML and it's improvements down the years.

автор: John D B

16 февр. 2020 г.

I enjoyed this course a lot. I found the lectures to be pitched at an appropriate level and were not boring or rudimentary. I would have liked there to be either links or optional lectures to the math and derivations for some of the algorithms that were glossed over in the lectures (such as backpropagation of neural nets) for those of us with the math background and interest, but I agree with the general approach of not bogging the class down with these full derivations. I think Andrew Ng's explanations/intuitive justifications of the various algorithms were what made the lectures really special. That intuitive understanding is much more valuable than going through a long derivation. I thought the programming assignments were excellent and impressed that the computer grading worked flawlessly. ( I used the MATLAB online option.)