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。)