Retour à Réseau de neurones et deep learning

étoiles

114,799 évaluations

In the first course of the Deep Learning Specialization, you will study the foundational concept of neural networks and deep learning.
By the end, you will be familiar with the significant technological trends driving the rise of deep learning; build, train, and apply fully connected deep neural networks; implement efficient (vectorized) neural networks; identify key parameters in a neural network’s architecture; and apply deep learning to your own applications.
The Deep Learning Specialization is our foundational program that will help you understand the capabilities, challenges, and consequences of deep learning and prepare you to participate in the development of leading-edge AI technology. It provides a pathway for you to gain the knowledge and skills to apply machine learning to your work, level up your technical career, and take the definitive step in the world of AI....

SS

26 nov. 2017

Fantastic introduction to deep NNs starting from the shallow case of logistic regression and generalizing across multiple layers. The material is very well structured and Dr. Ng is an amazing teacher.

GC

30 mai 2019

I have learnt a lot of tricks with numpy and I believe I have a better understanding of what a NN does. Now it does not look like a black box anymore. I look forward to see what's in the next courses!

Filtrer par :

par Mallikarjun C

•31 janv. 2019

I found this course to be extremely good. It covers nicely theory, implementation and application of Neural Networks and Deep Learning. Prof. Andrew Ng through his video lectures makes it fun and easy to learn this subject with the right emphasis on key points. The quiz's and program assignments are really good, reinforcing the concepts. In addition I found the Hero's of Deep learning conversation videos towards the end of each week, informative and thought provoking. This is my second course after taking Machine learning on Coursera. I am enjoying learning on Coursera. Thank you Prof Andrew and Team.

par Chatchai J

•27 nov. 2021

คอสนี้เหมาะกับใคร: ไม่เหมาะกับคนที่ไม่เคยเรียนเลย แบบมาเรียนอันนี้ ตายตายแน่ๆ เพราะเขาไม่ได้สอนแต่เริ่มต้น อารมณ์ประมาณว่าต้องมีความรู้ในตัวด้วยแล้วถึงมาเรียนได้ แนะนำคอส จาก อ. นพดล ช่วยได้เยอะเลย ควรมีความรู้ในการใช้ python -> pandas, google colab เบื้องต้นบ้างเพราะคอสนี้ไม่ได้สอนอะไรมาก มาปุ๊บจับเราโยนทำๆๆ แต่ยังดีเขายังไกด์ว่า search google ด้วยคำนี้นะแล้วลองอ่านดู ก็ถือว่าไม่ต้องไปค้นยันรากเง้า ความรู้แคลคูลัส กับ matrix ก็สำคัญ เพราะในนี้จะคณิตศาตร์พอสมควร (ใครบอกเขียนโปรแกรมไม่จำเป็นต้องเก่งคณิตนี้ไม่ใช่ชัวร์) แต่ถ้าไม่รู้ก็ไม่เป็นไรเพราะในคอสนี้เขาอธิบายพอสมควร แต่รู้ไว้จะสบายกว่า

par Maryllia K

•27 oct. 2020

Excellent step by step introduction to NN and DL. I took the course to brush up my ML/DL skills and strengthen my understanding on the DL basic concepts. I couldn't be more satisfied by the structure and overall layout of the course. The instructors give the right amount of detail for a beginners course without omitting important concepts. Plus, with the code available a practitioner can go ahead and practice the exercises on their own to make sure that they have mastered the concepts or identify the areas they might need to practice more. I really enjoyed the course and I would highly recommend it.

par Carsten W

•28 déc. 2019

Fantastic course with well structured Jupyter notebooks for your Python programming assignments. The assignments were pretty easy due to extensive explanations and repetition of key formulas from the lectures within the notebook. To be fair to others, maybe it was also a bit easy, because I just recently completed Andrew's older Machine Learning course (with programming in Octave and still highly recommended for a slightly deeper foundation in ML - I think), so I was already well familiar with the key concepts, vectorization etc, which I only had to transfer to Python. In any case, awesome course!

par Heshmat S

•26 déc. 2017

I've taken Andrew's "machine learning" course before, which I loved so much and learned a lot from it. The only issue with it was the use of "matlab/octave"; fortunately, he switched to "python" in this specialization course. :-)

This first course in the "deep learning specialization" is a very well though-out introduction to deep learning. Starting from logistic regression, Andrew builds upon the materials and masterfully introduces the more sophisticated concepts one after another. The programming assignments make the course even more fun and practical. Loved the course.

Thank you Andrew & Co. :-)

par Obaid S

•6 juil. 2019

This course is one of the best online course I have taken so far. With basic math knowledge (you just need to know what is a vector and what is a slope) you can complete all the assignments and the course itself. In this course, you get in-depth knowledge of how a neural network works by implementing it yourself. The best thing about this approach is that you will be very confident as you start playing with high-level libraries like tensorflow, since you will know what is going on under the hood. I think this course is a great place to start if you are new to deeplearning before using any library.

par Fabian A

•28 oct. 2017

I really enjoyed the Jupyter Notebook approach as it really suits my experience with Python3 and love of pedagogical and sound presentation of theory. The code can sometimes be a bit too forgiving in that it would be possible to go through it without thorough examinations of dimensions, calculations and the like. I, however, am doing this for learning rather than certification so it was a minor issue.

Really nice videos, a clear structure and a very thoughtful balance between the complexities of math and the "get things done" possibilities that jupyter notebook and Coursera permits. A great course!

par Debmalya M

•17 mai 2020

Perhaps this is the first course of this type that does not use any fancy python libraries to do something as complex as deep learning. It just uses numpy. For this reason, if tomorrow the python language gets obsolete, skill transfer would be very easy. The assignments are not too hard If you watch the videos regularly, but the contents are by no means easy to understand, particularly the parts where the instructor teaches matrix dimensions and backpropagation. I think watching the videos is not enough unless you practice the concepts yourself, with datasets downloaded from some other websites.

par critics

•14 mai 2018

This course is friendly to novice because Andrew is adept at making the originally complicated lessons easy to inteprete, and his clear pronounciation and moderate speed help students catch up his pace without extra effort even for non-native English Speaker.More importantly, we all known that Andrew is known as a prominent AI scholar around the world, and his intelligence is sparkling through the course, for example, the systematic course structure reveals his in-depth knowledge, as well as the practical advices on buliding a deep learning model shows his rich experience in actual implemention.

par Sikang B

•4 déc. 2017

Compare to the machine learning class years ago, this revamped NN and DL class took very modern approach and really take machine learning education to the next level by using new technologies, better programming models and last but not least, Python Notebook for education.

Assignments are helpfully guided, however the guidance felt a bit too excessive at times. Some text could be better delivered as hints rather than instructions.

This course is less demanding and is definitely perfect as an introduction course. The interviews are super relevant and highly engaging. Make sure you don't miss them.

par José A

•15 sept. 2017

It's only my first week in the course, and I'd say it's been good. It can be a little bit tedious to catch up with the terminology if you haven't seen any Data Mining or Machine Learning . Nothing that a good devotion of Google and YouTube-fu can't tackle.

Other than that, I have a very basic knowledge in the topic and I have had to do some good research about it. The 2nd week's lectures goes through each of the steps in building a Neural Network, including the explanation of a Gradient Descent, Logistic Regression, and derivatives.

I'll see if I can update the review after finishing the course.

par Paulo A F

•26 août 2017

Andrew Ng is the best! Congratulations to all the team involved in the course. It i at a very good level for everybody to join in. As an experienced programmer I though the programming assignments were on the easy side, but I guess they are at the right level for people coming from other areas. As for the maths, I think is a good idea to leave the deep stuff out of it and get people building the NN as a long as the maths behind it is solid, which it is in this course. People can delve deeper in other sources. I'm quite excited for the next 4 courses! All the best to the team and to the students!

par Mark D

•22 mars 2018

I loved the programming assignments. The tasks are nice, the visualization of the neural networks' decision boundaries is very helpful, and the setup with the Jupyter notebooks is just awesome. In other courses it is often required to set up a programming environment first, which sometimes takes more time than the programming exercises themselves. In this course it was possible to dive into Python and Numpy immediately - without worrying about file paths, environment variables, compatibility issues and other nuisances. The lectures were also very good. All the concepts were explained very well.

par Rahul K

•27 févr. 2018

Beautifully structured course! Feels like a walk in the park if you've already completed the 'Holy Bible of ML', i.e., Andrew Ng's Machine Learning course on Coursera.

Very good programming guidelines, and a gentle introduction to anyone who isn't aware of the core concepts on Machine learning.

If you're wondering whether you should complete the Machine Learning course first, by all means, go ahead. However, I can guarantee that there will be no hardships faced even if you're a beginner in ML and want to dive head-first into Deep Learning.

After all, it's Andrew Ng who we're talking about here! :)

par Omair M

•25 nov. 2017

Prof. Andrew Ng explains all concepts from a very fundamental level and even nervous students will feel encouraged by his insistence on "don't worry about it" for derivations you don't understand. The assignments have a lot of hand-holding but I needed that to focus on other more important concepts instead of debugging python code which can be learned in a different course. Overall, I have learned how to build a deep neural network using a building-block approach and gained confidence regarding this domain which I had previously taken to be mysterious and cryptic and perhaps for the elite only.

par Somnath M

•17 mai 2020

I had always wanted, formulae on the research papers to make sense in real world applications. However, as a novice programmer I wasn't been able to put those formulas into code and had to always go through multiple links and videos to make it working which was really a bottleneck as I didn't knew where to start. This course is really comprehensive and well crafted to make one understand the very basics to build a Neural Network and use them any Deep Learning Requirements. If you have a intermediate Python understanding, than no other course can help you create your own Neural Net. Thank You!

par MANUEL A F C

•31 déc. 2019

El primer curso de la especialización no solo te presenta el aprendizaje profundo de forma teórica sino que se ve reforzada con los ejercicios elaborados en el lenguaje python. Terminando con una construcción guiada de una red neuronal de 4 capas y entendiendo cada paso pues son planteados adecuadamente en funciones definidas con anterioridad. Recomiendo ver cada video a detalle y tomar apuntes, así como, practicar uno mismo implentando las funciones y decifrar que es lo que realiza cada línea de código desde un inicio para no perderse después (recomendación: lean los foros). Excelente curso.

par Sahil M

•20 août 2018

Andrew sir introduces the idea of neural networks using a single neuron(logistic regression) and slowly adding complexity — more neurons and layers. By the end of the 4 weeks(course 1), we are introduced to all the core ideas required to build a dense neural network such as cost/loss functions, learning iteratively using gradient descent and vectorized parallel python(numpy) implementations.

Andrew patiently explains the requisite math and programming concepts in a carefully planned order and a well regulated pace suitable for learners who could be rusty in math/coding. I love this course.

par Ye W

•28 oct. 2017

This course serves as a great intro. I saw many comments complaining that the course is a bit too easy. As a stats PhD student, I admit that the technical details in this initial course is trivial, but I feel that I learned a lot of useful things, e.g., vectorization, intuition, etc. In fact, the entire concept of deep neural net is very straightforward, i.e., nothing but a generalized linear model (GLM) from a statistical prospective. I feel what is important is the intuition behind it and how to implement it efficiently in practice. This course covers both aspects in great details, love it!

par GAUTHAM M N

•6 oct. 2020

It is a fantastic course for any one craving about the thrills of programming or math. Anyone without prior knowledge of programming can learn to like this course a lot. Knowing a bit basic math about matrices and calculus kinda makes it more fun, but no compulsorily needed. I am enrolled to all 5 courses in this specialization and will complete all of THEM! Andrew Ng is simply too good :)

P.s: It will get tricky a bit, because the 'what' of deep learning is tough to contemplate. Dont worry too much and just keep learning the 'how'. As time progresses you'll be able to understand it finally.

par Qiongxue S

•14 déc. 2018

This course helps me to understand what is neural network and how would we use the NN and deep learning method to solve the practical problem. This is real science. For the content I have to say that this is the best AI course I have ever had. The related theory and mathematic equations are all clearly explained. Besides I learned a lot from every assignment. The point to build NN is make sure we understand the theory first then the programming part will not be hard. But before learning the course I think we need to have basic knowledge about python. Excellent course! Thank you so much.

par Chris R

•9 sept. 2017

I have already completed Andrew Ng's Stanfor Machine Learning course on Coursera, but the neural network coverage was limited. This course helped me understand the underlying principles of deep learning more completely and I'll be taking all five to earn the specialization. The pace of this course seemed perfect for me having some knowledge of Python, linear algebra, and calculus. This course also helped to refresh older memories and learn new things about Python libraries like numpy. This is an excellent course and has left me very excited about possible applications for deep learning.

par Balwinder C K

•9 nov. 2020

Simply Awesome. I took another course from Andrew Ng (Machine Learning) 4-5 months back but didn't get much idea what's going on. Then i planned to take this deep learning course but before that i did quite to grasp the concepts. I must have watched the other course's video 50 times and must have done 50 small ML examples but that was beginning. This course was breeze as I knew this time the terminology, the concepts and specially what i wanted to learn from the course. It's always good to know underlying concepts as it give you power to debug the not so subtle scenarios.

Thank you Andrew

par Khizr M K

•18 avr. 2020

The course is very good and I found it really easy because I was familiar with python. There are two things which I want to suggest in your courses .

The first thing is you should also teach how to use python libraries for deep learning. This will teach students how to use library in different types of problems.

The second thing which I found was the course was bit easy and it should be made little bit difficult by removing certain hints such as formulae. This will force students to make notes seriously while listening your video lectures and implement formulae in their code on their own.

par Vu L

•20 août 2017

I took the ML matlab version that Prof Ng created, but could not make it because I could not understand the homework problems and the content of the course. Thus, I got out after the fifth week because I could not understand how to do the assignment. However, luckily, right after I got out, he opened this course. It was a relief that I was able to understand everything I did not understand from the previous one, and I was able to do the homework. Therefore, I would suggest this course to everyone who wants to learn AI. Thank you Prof Ng and your dedicated team for this tremendous effort!

- Analyste de données Google
- Gestion de projet Google
- Conception d'expérience utilisateur Google
- Google IT Support
- Science des données IBM
- Analyste de données d'IBM
- Analyse des données IBM avec Excel et R
- Analyste de cybersécurité d'IBM
- Ingénierie des données IBM
- Développeur(euse) Cloud Full Stack IBM
- Marketing appliqué au réseau social Facebook
- Analyse marketing sur Facebook
- Sales Development Representative Salesforce
- Opérations de ventes Salesforce
- Connaître la comptabilité sur le bout des doigts
- Préparation à la certification Google Cloud : architecte de Cloud
- Préparation à la certification Google Cloud : ingénieur(e) en données sur Cloud
- Lancez votre carrière
- Préparez-vous pour obtenir un certificat
- Faire progresser votre carrière

- cours gratuits
- Apprendre une langue
- python
- Java
- conception web
- SQL
- Cursos Gratis
- Microsoft Excel
- Gestion de projet
- Cybersécurité
- Ressources humaines
- Cours gratuits en Science de données
- parler anglais
- Rédaction de contenu
- Développement Web Full Stack
- Intelligence artificielle
- Programmation en C
- Compétences en communication
- Blockchain
- Voir tous les cours

- Compétences pour les équipes en charge de la science de données
- Prise de décisions basées sur les données
- Compétences en génie logiciel
- Compétences personnelles pour les équipes d'ingénieurs
- Compétences en gestion
- Compétences en marketing
- Compétences pour les équipes en charge des ventes
- Compétences en gestion de produits
- Compétences en finance
- Cours populaires de science des données au Royaume-Uni
- Beliebte Technologiekurse in Deutschland
- Certifications populaires en cybersécurité
- Certifications populaires en informatique
- Certifications SQL populaires
- Guide de carrière de responsable marketing
- Guide de carrière de chef de projet
- Compétences de programmation en Python
- Guide de carrière de développeur Web
- Compétences d'analyste de données
- Compétences pour un concepteur UX

- Certificats MasterTrack®
- Certificats Professionnels
- Certificats d'université
- MBA & diplômes commerciaux
- Diplômes en science des données
- Diplômes en informatique
- Diplômes en analyse des données
- Diplômes de santé publique
- Diplômes en sciences sociales
- Diplômes en gestion
- Diplômes des meilleures universités européennes
- Masters
- Licences
- Diplôme avec un Parcours de performance
- Cours de BSc
- Qu'est-ce qu'une licence ?
- Combien de temps dure un Master ?
- Un MBA en ligne vaut-il le coup ?
- 7 façons de payer ses études supérieures
- Voir tous les certificats