TCN
TCN copied to clipboard
How should I choose correct layers number?
Suppose my input sequences are of varying lengths (between 15 ~ 400) and 3 features.
What should I pass to TemporalConvNet(3, [3] * layers) for layers?