为什么Graves-LSTM Cell在deeplearning4j中有4 + 7 = 11个重量?

时间:2016-10-28 20:37:26

标签: neural-network lstm deeplearning4j

为什么Graves-LSTM图层Cell有11个重量,它们的用途是什么

以下示例可以生成权重列表:

{{1}}

出:

  

Weigths:

     

{0_W = [0.00,0.00,0.00,0.00],0_RW = [0.00,0.00,0.00,0.00,   0.00,0.00,0.00],0_b = [0.00,1.00,0.00,0.00]}

     

(11 weigth + 4 bias)

相比之下,输出使用DenseLayer 而不是GravesLSTM:

  

重量:{0_W = 0.00,0_b = 0.00}

     

(1个重量+ 1个偏见,这很清楚。)

0 个答案:

没有答案