Tian Z Zhou

US Patent:

20200234112, Jul 23, 2020

Filed:

Apr 10, 2019

Appl. No.:

16/380813

Inventors:

- Santa Clara CA, US
Tian Zhou - Menlo Park CA, US
Qun Gu - San Jose CA, US

International Classification:

G06N 3/08
G06F 7/544

Abstract:

A method of adaptive quantization for a convolutional neural network, includes at least one of receiving an acceptable model accuracy, determining a float value multiply accumulate for the layer based on a float value weight and a float value input, quantizing the float value weight at multiple weight quantization precisions, quantizing the float value input at multiple input quantization precisions, determining a multiply accumulate at multiple multiply accumulate quantization precisions based on the weight quantization precisions and the input quantization precisions, determining multiple quantization errors based on differences between the float value multiply accumulate and the multiple multiply accumulate quantization precisions and selecting one of the multiple weight quantization precisions, one of the multiple input quantization precisions and one of the multiple multiply accumulate quantization precisions based on the predetermined acceptable model accuracy and the multiple quantization errors.