A really good course that builds up the knowledge over the concepts covered in Course 1. All the ideas are applicable in real world scenario and this is what makes the course that much more valuable!
great introductory stuff, great way to keep in touch with tensorflow's new tools, and the instructor is absolutely phenomenal. love the enthusiasm and the interactions with andrew are a joy to watch.
автор: Rajesh R•
Great course to learn newer aspects of TF. For me a great revision of ConvNets and a confidence builder. If there's one thing I'd fix, it would be the autograder and how often it crashes.
автор: Amit G•
I liked this course, and also the way progression is taking place but the time required to complete this course needs to be reevaluated. This course can be finished in 8-10 hours easily.
автор: Amit K D•
Found the hands-on not very interesting. Couple of them focussed on file handling and stuff rather than on more important stuff that getting into the hoods of transfer learning, etc.
автор: Rakesh G•
I think this was a good course but the standard of exercises and quizzes was too easy. More conceptual questions especially in quizzes would help in understanding the topic better
автор: ashish s•
Overall good. Could have gone in bit more depth on how various hyper parameter tuning and regularization methods impact the model training. Provide some best practices tips .
автор: Gerardo S•
the last exercise needed a big upload, made it imposible (for me) to do. This was a problem not related to the subject, should use data downloadable directly from internet.
автор: Sokratis A•
Most of the Programming Assignments are a copy-paste routine from colab notebooks provided
the last one was pretty challenging though ^_^
(Im experienced in Programming)
автор: Eric L•
This course only requires few hours of work and I would like to see more depth. The parts on image augmentation and transfer learning were pretty interesting though!
автор: Luciano C•
Curso muito legal, a única coisa que ficou um pouco abaixo do esperado foi o último exercício da quarta semana, não foi construído com o cuidado visto nos outros.
автор: Vishwanadha K V•
The assignments are not challenging enough. The concepts are really well explained and for someone with no background in this area, this is a great learning asset
автор: Dimitry I•
Very good course that teaches you basics of convolutions, augmentation, transfer learning. Thank you to Mr. Moroney and the Coursera team for making it available.
автор: vaibhav t•
The course was good. The only problem was the last assignment where some of the functions went missing. It was difficult for a beginner to catch such glitch
автор: chaitanya m•
The best course to do. Especially after the specialization course from Andrew. It is really helpful to code all the concepts you learned from Andrew course.
автор: Ujjwal G•
I think most much of the course conent was same as the first course, this course could have been a little more advanced. But overall a great place to start.
автор: Moritz R•
Very nice the step from the first course was really nice. The achievements were harder to reach an all over the cose was less buggy than the first one. :)
автор: Estefania T•
The contents are a bit light from my point of view. I get it is to be accessible for more people but math and explanations are in some cases important
автор: Toqa A M E•
it was great course but I need some more details and the speaking was a little difficult as some of words are slang and the translation was so bad
автор: Subham S•
The course content was quite good and overall understandable but the exercises and quizzes were quite easy, they could have been more challenging
автор: Pray S•
The hand sign assignment need more explanation about using flow object from ImageDataGenerator since I just know only flow_from_directory object
автор: DING T K•
Learn a lot for CNN in this course, but require advance knowledge in Python & Numpy to understand the code as it was not explain in the course.
Exercise_4_Multi_class_classifier_Question-FINAL has problem if you entirely follow the tips, you can find the correct code in the forums.
автор: Gianluca T•
Very nice and interesting videos, cool concepts, amazing datasets. Exercises lack sometimes clear objectives, or provide unclear feedbacks
автор: Rodolfo V d A•
I guess one thing was not studied, the method .flow() which get the images generated by keras with the dataset labeled for the final test.
автор: Shubham S•
Lectures videos are amazing but the only problem in programming assignments. Programming assignments should have been properly designed..
More detail videos or links,examples for important techniques like dropouts and for like multi class classification which may be optional