Chevron Left
Retour à Apprentissage automatique

Avis et commentaires pour d'étudiants pour Apprentissage automatique par Université de Stanford

4.9
étoiles
166,828 évaluations
42,707 avis

À propos du cours

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....

Meilleurs avis

WZ
2 avr. 2018

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.

AF
16 mars 2021

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.

Filtrer par :

326 - 350 sur 10,000 Avis pour Apprentissage automatique

par Arvind G

13 août 2020

I was able to finish the course material without a solid grasp of the basics. The assignments did not reflect conceptual understanding. For example, I don't have a precise idea of how backpropagation works even though I implemented it in code. I just had to look up the pseudocode and translate it. I also did not understand SVMs and recommender systems with precision (how they compare and contrast with other algorithms). I think it will be better if larger portion of the code is expected to be written by the student.

par Ranjit B

24 déc. 2020

While the contents are good and the teaching pace is just right, I am deeply disappointed by the lethargy of Coursera in not fixing trivial errors in its assessment tests. Answers for even some trivial questions are graded as incorrect. Those result in incorrect grading and a frustration. When I am paying to get the assessments and a completion certificate, this is just NOT acceptable!

par Deleted A

13 juin 2020

Sound clarity is so poor sometime the volume is very low and some point it too hight, how can we concentrate on the course. Online course are stand on two main pillar video and audio, video s good but audio 2/5.

par Anton

11 mai 2018

Material of this course could be presented much deeper. Mr. Ng tries to avoid mathematical explanations.

par Timothy B

18 juil. 2020

Out of date, and video quality bad enough to be distracting

par Loftur e

17 sept. 2018

Assignments are very messy.

par pat

15 févr. 2021

I'm glad I didn't pay for this one. The answers in the quizzes are not correct. I checked them. Also, they don't tell you until week 2 that you will not be able to use any strings in your files in Matlab and Octave, everything has to be a number. I'm not sure this is useful to anyone. Bec the answers are wrong in the tests you won't be able to pass any of the quizzes, I got 60% on them. I retook each of them 6 times. I even checked the answers on Octave. Whoever wrote the quizzes did a poor job. I also did not understand any of the homework labs. I tried doing them and there were no instructions and the scripts did not work. Unfortunately, I can't recommend this class. It looks like the person who did the videos spent a long time on them, but whoever wrote the quizzes and homework did not check anything. Really sad. They could have made some money off of this one. Just sloppy.

par D M

14 juil. 2021

Has a lot of content, but just like you would experience in a university, the delivery comes in the form of:

1. Instructor talks at you for many hours

2. Now go take a test and see how much of what the instructor said stuck.

The course does very little to encourage understanding and comprehension of the material, so if you actually want to walk away with the ability to apply the material that has been presented, you are going to have to look for resources outside this course to complete your understanding.

A​lso, the homework problems frequently feel like they are from an entirely different course. Referring back to the videos for help offers little to no help in understanding what is desired in the homework.

par Richard L

5 oct. 2021

​I have tried three times to purchase this course unsuccessfully. My credit card is valid and It works for other purchases. Coursera customer support is totally unhelpful. Coursera should treat its paying customers better. I am a subscriber to Coursera Plus and before that I paid for a number of courses. Up to this point, I had been reasonably pleased with Coursera.

par Sabahat

13 févr. 2021

In the beginning there are notes to explain each video. In the last few videos, there are no notes and it becomes impossible to keep pace with what the instructor is saying as the slides also don't mention the key points that one is interrogated upon in the quiz. The assignments are also extremely tedious and I at least did not learn much from them.

par Maarten d s

7 janv. 2020

the quizzes were very good but the programming tests were badly made and not well enough explained.

some problems can come from having Dutch as first language others from the continuous task of just translating the formula given into a formula for the programming. or just plain old copy paste from the instructions of the file itself

par James L

6 févr. 2021

I sleep every single time when I am watching the webinar only for 5-10mins.

Need more visual aids and examples. Also the voice is so calm, nothing exciting to learn.

If you want to fall sleep fast, I recommend you watch the videos.

Unenrolled.

par Miguel C C

6 juil. 2020

Lioso y muy mal organizado. Las preguntas de los test hacen referencia a otros temas y la puntuación es injusta. En general, muy decepcionado y voy a pedir la devolución del dinero.

par Gosforth

10 juil. 2019

My feeling is that the author of this course has no idea what is "Machine learning" - I have the impression that he repeats slogans which he does not understand.

par Lorenzo V

23 mai 2019

No math, purely intuition and drive through formulas not demonstrated. You can't improve after this course because you don't really know why you did what you did

par Golam R

24 juil. 2021

​The course is designed based on Matlab/Octave. But Python is more intuitive language for this field. So i lost interest on this course.

par Abdullah D K

18 févr. 2021

This is not a course, more like listening to the people who talks about machine learning and then writing your feelings about them.

par Romie C M

8 juin 2020

A good set of questions contain only one best answer and that is in measurement and evaluation.

par Uri Z

9 sept. 2016

Very basic and superficial course. Apologies each time derivatives need to be used.

par Ruslan Z

23 oct. 2020

theory is intuitive and ok but rated program assignments are just waste of time.

par Rishi A

4 déc. 2019

Locked assignments are really frustrating.Why to wait till a specific date?

par Siddharth K

1 avr. 2020

Python should have been great language for this course.

par Vivek P

4 oct. 2021

Course not updated in ages.

par Aly E

10 juin 2021

I have to say Andrew did a pretty wonderful job in this course. I was a person with a very nice software development experience but never had to deal with machine learning. The last time I had to deal with calculus, algebra or mathematics in general was about 7 years ago (in Arabic, and having to deal with that in English is another story), thus I had approximately zero mathematics knowledge. Before this course, I attempted different approaches into this field but throughout them, I would either fall in a valley of philosophy or I would have to stop every few minutes and check the mathematics behind what's just happened.

The way Andrew approached the content in this course makes perfect sense to me (and I assume, to anyone with similar background). He's not the kind of teacher who'd plot complicated things onto the board and tells you that you should use it, instead, he would build the components of everything bit by bit until it makes perfect sense. He also has a good estimate of how hard/complicated something might be/seem to new comers and thus he instructs you throughout the course to be gentle on yourself if you don't get it at first.

Also, the vast majority of quizzes and programming assignments in this course put you in situations where you have to deal with tricky confusions in order to work things out and thus try to make sure that you have a deep understanding of what's going on.

I also like the quality of the content provided in this course. Andrew didn't just tell you "hey, here're the algorithms and that's how you use them, go use them", instead, he dedicated a decent amount of effort trying to explain how to choose which algorithm and when and why, and how to "not depend on gut feeling" but instead diagnose and debug different situations you might find yourself in.

Judging by earlier approaches I attempted before this course, I believe that it might've taken me a very long time to obtain the knowledge provided in this course.

One minor draw-back of this course is that unlike the first half, the last few weeks don't have reading recap after each video session. Another one might be the fact that the weight of this course (in terms of time and effort needed to complete something) is not equally distributed across the weeks (one programming assignment took me almost two weeks to complete, and two weeks in the course took me one day to complete).

par Augustin L K

22 sept. 2021

(1) I​ think the lecture material has to be revised: My suggestion is this: to make the lecture clear use 3 by 4 matrices to describe or explain each steps ( this will ensure that the student can solve the problem manual and therefore use any language to code). The summation over i, j, k with the training example m ; makes less pratical. there are many repetition which can be removed as well ;

=​> Based on my coding experience, it is very good to keep the mathematical generality aside for clarity purposes; making the lecture very easy ; Really there is a total disconnection between the developped theory ( equations) and their coding ; it makes it hadrer ;

I​f one can solve or translate the mathematical equation manually then the programming assignment will be easier

(​2) you start each topic in a wonderful way by providing sweet examples ; however , in this case, there is not a general formulation of the example that will help easily the student decided about the next example different from what you gave: In order words : when you talked about aircraft engine on anomalous algo; you did not clearly (in words) say what are the characteristics of a problem which will be identical to the aircraft exaple

(​3) About the assignment, there is a huge dark side : the student has to complete only whatever it is asked to; however in practice the student will have to developped more than what he is asked as assignment, It is therefore important to present fistly important to tell the student if there are some libraries developped in the current language used in the lecture ( in this case Matlab) : in other words:

l​et say , the in the << exercise>>, the main code is ex3 where there are lines of cod ( which the student will have to write in practical scenario as well : This ex3 is not concern by the assignment, which is good BUT at least say a word about this: Let the student know if in practice he will have the task to write this main file ( ex3), will be better

T​hanks , your lecture is good in the sense that it connect two main sides upon which the ML is made : mathematical side ( by this I mean the useful resulting equation to be used ) and the pratical formulation and application of this in an algorithm

T​HANKS !!!