Machine Learning with TensorFlow

t’s have been two years since the publication of the first edition of Machine Learning with TensorFlow. Two years is a long time in the field of artificial intelligence. Today, we are fascinated by a human-language model with more than 80 billion artificial neurons that have learned more than 170 billion parameters. The cost of training such a model is measured in millions of dollars. Lex Fridman of MIT projected that with the improvements in computing and algorithm design, we’ll soon train a model the size of the human brain for less than a few thousand dollars. Just think””in our near-term future, we’ll train an AI model with the raw capacity of the human brain for less than the cost of a Peloton stationary bike. Writing a book to capture this fast-moving technology is fraught with risk. By the time Chris wrote a few chapters, researchers likely produced newer, more elegant approaches to solving the same problems. Yet there are perhaps only 10,000 people today who understand AI deeply. You want to jump in, learn, and start using AI in your work