Closed narcise closed 4 years ago
would you help me to find which part of the code is about Fast inference by caching intermediate states in convolutions?
Please see https://github.com/r9y9/wavenet_vocoder/blob/a835a8eebfec5e049d93f629c4ee2f1daab56329/wavenet_vocoder/conv.py#L17-L46.
self.input_buffer is the intermediate state.
self.input_buffer
would you help me to find which part of the code is about Fast inference by caching intermediate states in convolutions?