In this chapter, we introduced the Caffe deep learning framework and examined the relationship between Caffe and Caffe2. We examined the Caffe and Caffe2 model file formats. Using AlexNet as an example network, we looked at how to convert a Caffe model to Caffe2 format. Finally, we looked at the difficulties in converting a Caffe2 model to Caffe.
Caffe is a DL framework that has reached its end of life and no new features are being added to it. In the next chapter, we will look at contemporary DL frameworks, such as TensorFlow and PyTorch, and see how we can exchange models to and from Caffe2 and these other frameworks.
