Someone deeply recommended me this Coursera course https://www.coursera.org/learn/neural-networks-deep-learning/ , for me it turned out to be quite disappointing, the lessons show lot of formulas with hardly any example, they don't communicate passion and interest, and they are totally in the wrong order in relation to the practical tests, so you have to take a test before the theory has been explained.
The exercises are useful, but you can find them here and you can practice in here https://jupyter.org/try .
Some exercises are also available here https://github.com/Kulbear/deep-learning-coursera/blob/master/Neural%20Networks%20and%20Deep%20Learning/Week%203%20Quiz%20-%20Shallow%20Neural%20Networks.md
Someone recommended me a textbook Deep Learning by Ian Goodfellow, inventor of generative networks. But it is very theoretical. There's a very nice course on YouTube from NYU and it's by Alfredo Canziani and Yann LeCun who's the guy who invented convolutional networks. There is a channel called Deep Lizard which has a course on PyTorch which is a framework to make neural networks.
The exercises are useful, but you can find them here and you can practice in here https://jupyter.org/try .
Some exercises are also available here https://github.com/Kulbear/deep-learning-coursera/blob/master/Neural%20Networks%20and%20Deep%20Learning/Week%203%20Quiz%20-%20Shallow%20Neural%20Networks.md
Someone recommended me a textbook Deep Learning by Ian Goodfellow, inventor of generative networks. But it is very theoretical. There's a very nice course on YouTube from NYU and it's by Alfredo Canziani and Yann LeCun who's the guy who invented convolutional networks. There is a channel called Deep Lizard which has a course on PyTorch which is a framework to make neural networks.