tf.layers.conv1d(dilations)
JamesX2015 opened this issue · comments
I'm wondering why we use time_to_batch + causal_conv + batch_to_time in ops.py, rather than tf.layers.conv1d with dilation_rate>1. Is it because it computes faster or old versions of tf don't support?
I simplified the code using tf.layers.conv1d.
See the Issue