April 2017
Intermediate to advanced
320 pages
7h 46m
English
The way TensorFlow is going to increase the speed of its programs and incorporate more devices that can run TensorFlow is with this JIT compilation via XLA.
The way XLA is working is summarized in the following figure:

TensorFlow team has developed this compiler infrastructure such that you can hand it a TensorFlow graph and then an optimized and specialized assembly comes out.
This is the compiler infrastructure that TensorFlow team have developed such that it can take a TensorFlow graph ...
Read now
Unlock full access