As I talked on this year's resolution, I'm going to learn about reinforcement learning. However, I have never learned about deep learning, so first dipped my toes in the water of the mystery.
So I first tried this book - TensorFlow Machine Learning Cookbook - but honestly this is terribly written. It seems this book has never been reviewed in advance; still in 1st chapter, but due to lots of typos and not that good English writing, I ceased to read the book.
Some variables are called even though they have not been declared, others are just complete messes like
linear_tsr = tf.linspace(start=0, stop=1, start=3), and awful English writing such as the following (excerpt from 1st chapter):
Note, that since these tensors depend on prior tensors, we must initialize them in order. Attempting to initialize all the tensors all at once willwould result in an error.
Now I go on to try another book. Unfortunately, I don't recommend this book...