Depth learning model matrix compression method and device

Depth learning model matrix compression method and device

  • CN 105,184,369 A
  • Filed: 09/08/2015
  • Published: 12/23/2015
  • Est. Priority Date: 09/08/2015
  • Status: Active Application
First Claim
Patent Images

1. , for a matrix compression method for degree of depth learning model, last layer of linear layer of wherein said degree of depth learning model connects M hidden node and N number of class node, the weight matrix of described last layer of linear layer W=w11w1N...wM1wMN,Described method comprises:

  • Step S101;

    according to the absolute value of the element of described weight matrix W, calculating K value;

    AndStep S102;

    described last layer of linear layer is decomposed into the first linear layer and the second linear layer, the weight matrix of wherein said first linear layer is the matrix of M*K P=p11p1K...pM1pMK,The weight matrix of described second linear layer is the matrix of K*N Q=q11q1N...qK1qKN,The output of described first linear layer is the input of described second linear layer, and M*N>

    K* (M+N), so that described weight matrix W is compressed.

View all claims
    ×
    ×

    Thank you for your feedback

    ×
    ×