This chapter is the pinnacle of the hardware part of our journey. We are now ready to take all the chips that we’ve built in chapters 1–3 and integrate them into a general-purpose computer system capable of running programs written in the machine language presented in chapter 4.
How to use transfer learning in keras
The difference between accuracy for Cats vs Dogs with data augmentation and without
Multiprocessing.Pool中 apply, apply_async, map, map_async的区别；多进程文件写入解决冲突。
Problem fixed and skills gained
Recognizing real images of Cats and Dogs
Set http proxy for package control;Use the sublime text3 to connect the remote server; Gist and sublime text3 integration
Some memory elements built from sequential chips.
The training set is the data that is used to tell the neural network model that 'this is what a horse looks like', 'this is what a human looks like' etc.
Exploring how Convolutions and Pooling work