6. Intel64 instruction set extension
The Intel64 instruction set, with its variable-size format, makes it easy to add new instructions. Under the heading "Deep learning Boost", a number of instructions have been defined to speed up AI code, particularly for convolutions.
The calculation of a convolution point was presented in § 5.1 . It is of the style S1 = F1.E1+F2.E2+F3.E3+...+Fi.Ei+..., i.e. a sequence of multiplications-accumulations....
Exclusive to subscribers. 97% yet to be discovered!
Already subscribed? Log in!
Intel64 instruction set extension
Specialized operators: tensors and tensor cores
Article included in this offer
"Software technologies and System architectures"
(
230 articles
)
Updated and enriched with articles validated by our scientific committees
A set of exclusive tools to complement the resources
Bibliography
- (1) - NIELSEN (M.) - Neural Network and Deep Learning, - http://neuralnetworksanddeeplearning.com/
- (2) - TensorFlow - -
Exclusive to subscribers. 97% yet to be discovered!
Already subscribed? Log in!
