Great Course Overall\n\nOne thing is that some videos are not edited properly so Andrew repeats the same thing, again and again, other than that great and simple explanation of such complicated tasks.
This is very intensive and wonderful course on CNN. No other course in the MOOC world can be compared to this course's capability of simplifying complex concepts and visualizing them to get intuition.
by Adrien S•
Great overall course, keep teaching please ! I learnt a lot. I have a Ms degree in Machine Learning but we didnt had the time to really learn about Deep Learning. I feel it was a great introduction to the field and I feel confortable now to get more in details about everything and read papers etc.
So thanks for that, and I can't wait for part 5 about RNN
by P M K•
This was a really good course to see mini projects getting executed. It gave quite a lot of practical insights working on the problems. The only issue was that week 4 assignments had some bugs in code comments due to which people spend quite a lot of time debugging causing unwanted waste of tine and frustration. Please correct the errors.
by yuji w•
nice program to learn about convolutional neural works. I always fascinated about convolutional networks and this course gives me the very nice introduction and sort of in-depth knowledge and first hand programming knowledge in this area. The instruction and nice and start from easy and slowly get you into the deep knowledge. Great course and nice work.
by Daniel C•
This course covers the basics of convolutional neural networks. After you understand the materials covered in this course, you'll know how smart phone cameras auto focus on faces. You'll also learn the basic building blocks that powers self-driving technology. These are just two of the many cool concepts you'll learn in this course. Highly recommended!
by Vishaal K M•
The programming exercises require much more attention than you think it does. Although it's required to only fill in the code in specific areas and not too much either, the foreword before each code section must be studied carefully if you are to build your own convnet. The video lectures are pretty straight forward, so there's nothing to worry about.
by Martín C•
Unos de los cursos más didácticos que he realizado. Muy claras las explicaciones de Andrew Ng sobre todo con respecto a las capas que componen una ConvNet. ¡Lo disfruté! Recomendado.
One of the most didactic courses I have ever taken. Andrew Ng's explanations are very clear, especially regarding the layers that make up a ConvNet. Enjoy it! Recommended.
by Cem O•
Just like the other courses in this series, this course was prepared with great care to optimize the learning outcome. Clear and motivating lectures, great selection of up-to-date methods and very illustrative examples. I would like to thank Prof. Andrew Ng and all the course staff most sincerely for designing and making available these great courses.
by Guangyu L•
Very good learning experience. Prof. Ng gave a lot of insights about not only the CNN frameworks but also some real world working experience and hints which were very informative. For this one , I had very heavy work load during learning, I recommend people take it in a continuous manner, this helps you understand and connect every knowledge nodes.
by Abhishek A•
Excellent Course!! By doing the this course I am now feeling very confident in CNN. This course is very important for all whether they may or may not work in CNN/images. This fundamental learnt here can be used in other domains of deep learning.
Thank you deeplearning.ai Team for proving this wonderful course. It has opened new opportunities for me.
by Pin Z•
This is a very good course to get to know the basic concepts of CNN and to start hands-on programming to implement CNN. Andrew's lecture gives very clear explanation of the principles of CNN, as well as introduction to state-of-the-art example network structures. The exercises help to build essential skills to program CNN using TensorFlow and Keras.
by Youssef H•
I have really learned a lot from taking this course. During the course you will be exposed to the state of art deep learning architectures by understanding the theory behind them in lectures and then you will get to implement them in the assignments. I have taken the first three courses and I think that definitely this course is by far the best one.
by Elidor V•
The course was simply great. It starts from the real basics of Convolutions, gives you all the needed theoretical background, then starts to focus on real-life scenarios. Also worth mentioning that is not a piece of cake. The given assignments are not easy in general, but after completing those the benefit will be more than clear. 100% recommended!
by Nazmus S E•
Although this course was a bit difficult compared to the previous one, it was more informative and taught a lot of real-life applications of CNN and Deep Learning. The assignments of Week3 and 4 involved pre-trained models. Explanations of them were not given but links to where we could learn about these models were given. Overall a great course.
by kumud C•
I was scared of CNN and thought that it's quite overwhelming to learn such new concepts like Residual Network, YOLO, Face recognition. This course helped me in understanding these algorithms intuitively and practically. I loved watching videos and will watch in the future as well to revise the concepts I learned. Thanks to Coursera and Andrew Ng.
by Hector L•
I enjoyed this course. I learned a lot about Convolutional Networks and the assignments were very fun to complete. The assignments are difficult enough to lay the groundwork for the subject - but you definitely need to take your time to understand and probably run experiments on your own.
I loved the ResNet, YOLO, and Face Recognition assignments.
by Yogesh C•
This course was amazing and interesting. The tutorials and quizzes were great. But I was looking for the implementation of CNN from scratch without using tensorflow.
Rest as mentioned this was an amazing course. Now, I have a better understanding of YOLO algorithm, face recognition, Neural style transfer. Thanks to Andrew and the rest of the team!
by Sadam H•
Learned some interesting concepts about different state-of-art ConvNets. Although I was hoping that in Face Recognition Programming exercise there would be some code implementation exercise or example about one-shot learning and Siamese network, it would have been perfect. Nonetheless, very nice structured course to learn intuitions intuitively.
by Abanoub A•
The Way Prof. Andrew explains things, taking us from simple stuff to the complex conclusions by ourselves making it so much easier and convincing!
The course content was great and assignments were fun, I like that in the end of each assignment there is always a cell that's like a "playing ground" allowing you try and test the models you created.
by Hardik V U•
This course is good from both the perspective: Research and Development. This course involves many real life applications which will help us to understand the real life problems and also will help in tacking such problems. So, I would strongly suggest to go for this course which builds the fundamental for computer vision and pattern recognition.
by Jatin s•
This was by far the most engaging and fun course in the entire specialisation.I guess as the concepts build up the tasks get more interesting and exciting.Their was a ton of content in this course , you need a very sound and solid background of the previous courses in this specialisation to get a firm hold of the concepts taught in this course.
by Okta F S•
This is very good course. From here you can learn so many things, start by learn basic convolutional operation, intro to some of ConvNet architecture like Inception, Residual block etc. And the most important thing is you can applied your knowledge to build some use case systems like object detection, neural style transfer, and face recognition
As a beginner I have learnt a lot of topics with good clarity. Assignments have given me good understanding of the topics learnt.
I think the assignments should some more difficult and students should be able to spend some more time understanding the code and writing code of their own.
Thank you very much for making learning affordable and easy.
by William v•
The libraries needed such as tensorflow, might need to better support (a special segment on them beyond the overview). Those models are complex and deep and using those libraries wasn't clear to me even though I managed to get the solutions, I needed time to study those libraries and they are rich and complex. I enjoyed the course immensely.
by Wanda L•
Fantastic course about Convolutional Neural Networks! For me the best part of the course (and the specialization, too) is the assignment. You could hardly find a similar friendly, supported and easy-to-follow homework elsewhere in the world, even in some universities. Thanks to Andrew, and thanks to all teaching assistants in the community!
by Eddy P•
All are pretty good! Except for the low speed while running the training process which I think have in fact hurt the course's completeness. Because we have skipped many important training processes and instead use pretrained models to save time. I suggest maybe we can collaborate with Google and put the programming assignments on the Colab.