Practical Computer Vision
上QQ阅读APP看书,第一时间看更新

Keras for deep learning 

Keras is a Python based API that uses TensorFlow, CNTK, or Theano as backend for deep learning. Due to its high level API and simplified abstraction, it has been quite popular in the deep learning community. We will be using this library to study CNNs. To install this, first install TensorFlow as described in previous section, and use the following:

pip install keras

There is no separate version for GPU. For installing specific versions of Keras, such as Version 2.1.2, use following: 

pip install keras==2.1.2

The latest version of Keras at the time of writing this book is 2.1.2. To check the version of installed Keras, use:

python -c "import keras;print(keras.__version__)"

If TensorFlow is installed from previous sections, it will use it as backend.

To use Keras, one of the prerequisites is basic knowledge of deep learning. In this book, we will see it in Chapter 5, Convolutional Neural Networks.