Temporal Convolutional Network
Causal Convolutions
灵活的感受视野
比如改变卷积核的参数,容易控制模型的记忆大小;
Receptive field
- Receptive field = nb_stacks_of_residuals_blocks * kernel_size * last_dilation.
- If a TCN has only one stack of residual blocks with a kernel size of 2 and dilations [1, 2, 4, 8], its receptive field is 2 * 1 * 8 = 16. The image below illustrates it:
KERAS-TCN
https://github.com/philipperemy/keras-tcn