Recently, TensorFlow Machine Learning Cookbook
has been published in Japan and I took one. Personally, TensorFlow has been very useful and also had been confusing for me. Because of its rapid growth, sometimes new functions come and other times those are integrated into the others. Its tutorial has big volume and although I know by reading it in good manner we can understand TensorFlow and use it well, it seems annoying to check those, one by one. Especially when you are new to machine learning and deep learning, it is just huge obstacle to read it all although in my opinion, TensorFlow isn’t the tool for the person who just started to tackle with those areas.
In a nutshell, TensorFlow is not so easy to study especially on its early steps of learning.
Simply, I wanted the TensorFlow document to have the contents of this book.(I know this hope is too much)
This book lets us learn about TensorFlow step by step. By following the book from the first page to the last page, you can review the basic knowledge of machine learning and be used to writing TensorFlow code.
The index is as followings.
- Introduction
- TensorFlow_Way
- Linear_Regression
- Support_Vector_Machines
- Nearest_Neighbor_Methods
- Neural_Networks
- Natural_Language_Processing
- Convolutional_Neural_Networks
- Recurrent_Neural_Networks
- Taking_TensorFlow_to_Production
- More_with_TensorFlow
As you can see, the contents have the wide range from simple linear regression to Convolutional Neural Network, Recurrent Neural Network and so on. Even Nearest Neighbor methods and some unsupervised learning methods are covered.
By many examples, we can learn how to use TensorFlow.
On the github, you can see the codes which are used on the book.
If you are new to TensorFlow, this book works as a part of tutorial and documents. If you are not, you can review the machine learning algorithms and how to use TensorFlow through many examples.