Skip to main content
. 2021 Feb 1;21(3):972. doi: 10.3390/s21030972

Table 2.

Number of trainable parameters.

Layer Type Output Shape (Timesteps × Features) Parameters
input1 InputLayer 1 × 4 0
lstm1 LSTM 1 × 16 1344
lstm2 LSTM 1 × 4 336
repeatvector1 RepeatVector 1 × 4 0
lstm3 LSTM 1 × 4 144
lstm4 LSTM 1 × 16 1344
timedistributed1 TimeDistributed 1 × 4 68
Total parameters: 3.236
Trainable parameters: 3.236
Non-trainable parameters: 0