Customising Your Models with TensorFlow 2
Imperial College London
I highly recommend this course to everyone who is willing to be using TensorFlow 2. I have been using TensorFlow for years, but I still learnec a lot from it. One caveat is that this course actually is not an easy one, it won’t teach your what machine learning is, it only teaches you how to use TensorFlow, so you have to obtain enough theoretical knowledge before getting started.
Previously, in its sister course, the Sequential API of Keras was taught, so naturally this course begins with the Functional API, that allows you to use multiple inputs / outputs, to build complicated non-linear topology and more. Functional API is essential for building recurrent neural networks (RNN) and residual neural networks, you will implement them later in the course.
The second part of the course is about dataset. The course covers the Keras datasets and their generators, and more importantly, the
tf.data.Dataset module. Be sure you know how to implement filtering, mapping, and the use of
take() methods. You will count them on to extract correct data for you in the final capstone project.
Next, the course shifts the focus to embedding layers, masking and recurrent neural networks. You might feel too sudden and anxious if you lack enough knowledge on sequential models like LSTM. Calm down, find some other resources to help you. The last part is the most subtle and challenging, you will be derive (subclassing) the
Layer classes directly and write your own training loop to completely unleash the power of TensorFlow.
Finally, I have to say the final capstone project is epic. You will implement a real English-German translator from scratch. I spent entire 3 days to complete this homework. Be cautious about the loss function you choose and the way you implement the weights in the custom layers. Unwise decisions will entrap yourself and it will be hard to debug.
Upon the completion, you will be very proud of yourself. Good job!
I am Kesler Zhu, thank you for visiting my website. Checkout more course reviews at https://KZHU.ai
All of your support will be used for maintenance of this site and more great content. I am humbled and grateful for your generosity. Thank you!
Don't forget to sign up newsletter, don't miss any chance to learn.
Or share what you've learned with friends!Tweet