I really enjoyed this course, it would be awesome to see al least one training example using GPU (maybe in Google Colab since not everyone owns one) so we could train the deepest networks from scratch
Great course. Easy to understand and with very synthetized information on the most relevant topics, even though some videos repeat information due to wrong edition, everything is still understandable.
교육 기관: keerthi k•
Thank you so much Coursera. I have been doing this specialization properly, but suddenly I had an accident which took almost 10 days to become normal. During those time several assignments were overdue, but Coursera extended their assignments deadline twice and helped me complete this course. So once again I thank Coursera.
교육 기관: Abhishek S•
The CNN is always found as one of the trickier concepts to follow and it was actually very hard for me to figure out what these Conv layers are doing. But this course is so robust and easy to follow that I was even able to read the research papers on advanced CNN architectures with relative ease. Thanks to Andrew and team.
교육 기관: ANSHUMAN S•
It has been a great journey through learning CNNs it was quite interesting rather than all other courses and I got to know really very new ideas which i can implement in my own models.
Once again I want to thanks Andrew Ng and all other teachers of Course
and a special thanks to Coursera for giving me this ample opportunity
교육 기관: Nick H•
Awesome course if you want to understand the basics of CNNs along with recent applications of these algorithmns.
As usual, both Andrew's material and his presentation style kept me both engaged and interested to a point that I got ahead of the weekly schedule...which is probably a good metric in terms of course success
교육 기관: Nikhil V K•
Great course by Andrew Ng sir. It gives us a great insight into many case of studies of state of the art ConvNet. Gives us a lot of intuition about object detection systems in autonomous driving and landmark detection , one shot learning for face recognition and a fun way of applying ConvNets for neural style transfer!
교육 기관: Wang F•
Despite the confusing bug and server running problem in the last assignment of happy house ,
the course is still great . Compare to the other three ones, it's the hardest course for me by now .
You may feel stuck in some practice questions and program .Worth spending time to review the
stuffs of the course again。
교육 기관: Pawan S S•
One of the best courses I found to learn convolutional neural networks as a beginner. All the subject matter are well structured and the flow of the module is very easy to follow and understand. Together with the programming assignments, I was able to quickly grab the essentials of CNN. I highly recommend this course.
교육 기관: Edson C•
This was the most difficult course I did in this specialization, but I loved it, I loved it very much. Thank you very much dr. Andrew and coursera for the opportunity, I really understand the importance of studying computer vision and this course was very useful in this journey. Thank you very much, I really loved ...
교육 기관: 杨建文•
The last 2 courses were delayed, but the positive side for me is that, in the beginning I proceed too fast and didn't learn that well, the delay made me take more time on such a valuable course, carefully reading and memorizing the instructions of assignments. I'm really grateful for Prof. Ng's excellent instructions.
교육 기관: Adam F•
I completed the entire specialization and having nothing but good things to say. Highly recommend it! Lectures are engaging, and Andrew does a fantastic job explaining some very complex topics. Programming assignments are challenging in a good way. You’ll really feel like you’ve learned a lot by the time you’re done.
교육 기관: Krishna M•
This Course was exceptional and upto mark. I learnt a lot of stuff easily and was able to implement into the real world example. This was really helpful for building up my resume. I thank Andrew Ng and Coursera team for giving financial aid to take up this course. The amount of knowledge gained is so valuable to me.
교육 기관: Eric C•
Awesome. This course was much more dense than the other ones, there is so many areas to review. Since this course is about my favorite subject, I will need to pause and rework on each individual points and associated papers (yolo, nst, similarity learning) which will probably take me weeks... Prof Andrew is the best
교육 기관: Arvind N•
I thoroughly enjoyed taking this course. Beautifully designed...Thank you!
I had written a detailed review of the first 3 deeplearing.ai courses at : https://medium.com/towards-data-science/thoughts-after-taking-the-deeplearning-ai-courses-8568f132153
I will review this CNN course as well, in the form of a blog post.
교육 기관: Benjamín V A•
Great course, provided many clear explanations I has been searching before. The one thing they could improve is that some of the practical exercises seem more focused in the framework than the algorithms. (I spent more time googling how to pass parameters to specific functions than actually using the algorithms)
교육 기관: Wade J•
Good amount of challenge for after work learning. Nice exposure to different applications of AI. Fun.
Andrew Ng is awesome at explaining the concepts. Almost anybody would be able to understand them after he presents them. I also appreciate how genuine he is. You can trust that there is merit to what he tells you.
교육 기관: Glenn P•
Another excellent course. Convolutional Neural Networks is no longer a mystery. I like the fact that Andrew doesn't teach this as an academic class but has a very practical approach that can be applied right way. He lets you know the strengths and weakness of each of the NN and gives his personal opinion as well.
교육 기관: Yijie•
It is a great course that covers most part of Convolutional Neural Networks. I have learned a lot from it. Thanks Andrew! Only one suggestion: we have learned dropout and the batch norm in previous courses. Because they are such important tricks, it would be better if you could cover how they can be used in CNN.
교육 기관: Ahmad B E•
Greatest cores for me till now on deep learning. I recommend it for deep learner or computer vision student. The best thing in this course is that it is very practical and up to date, and full of research papers of algorithms that Google and Facebook currently uses. Thanks a lot Prof Andrew Ng you are the best.
교육 기관: Yuri C•
What a ride! I am not even very much into Deep Computer Vision, but this course made me finally understand how tensors algebra works and how they flow in the network. Andrew is just able to put it in so simple terms and in a very accessible way that just for that the course is already very remarkable! Congrats!
교육 기관: Parab N S•
An Excellent Course to make people understand Convolutional Neural Networks in good depth and with ease. The detailed understanding of the major Convolutional models like YOLO and ResNet is like an icing on the cake. I would like to thank Professor Andrew N.G. and his team for developing this wonderful course.
교육 기관: Alejandro M•
Muy bueno para empezar a entender los conceptos de las capas convolucionales. Luego muestra modelos profundos como AlexNet, VGG16, ResNET, Inception que se pueden entrenar usando transfer learning. La parte de detección de objetos es la mas complicada. La parte que más me gusto fue la de reconocimiento facial.
교육 기관: Jeffrey T•
The intuition and examples made this course easy to understand and learn. I loved how Andrew decomposed current published papers into an easy to understand format. All of the important points to remember were highlighted without wasting time on the minutia. Thanks for all the hard work put into the course.
교육 기관: H A H•
I enjoyed a lot in this course...who wants to know how to build the CNN model...then this course is absolutely for them..they should try 100% this course. this course gives u insights into how to build your CNN model this one is I think the best course for that...thank u sir for this type of good content...
교육 기관: Carlos A L P•
Nice exploration of CNN theory covering theory and Python exercises through different algorithms. One recommendation would be update broken links and re-write comments in code as sometimes it is not clear what variable or what is needed to complete the required functionality, specially on ungraded exercises
교육 기관: MBOUOPDA M F•
This course explains the details of CNNs with a great simplicity. It also presents some state of the art CNN architectures with their ideas very clearly. Finally the assignments allow to implement several CNNs and also show how transfer learning is used to perform face recognition and neural style transfer.