|View source on GitHub|
Tile the batch dimension of a (possibly nested structure of) tensor(s)
tfa.seq2seq.tile_batch( t, multiplier, name=None )
For each tensor t in a (possibly nested structure) of tensors,
this function takes a tensor t shaped
[batch_size, s0, s1, ...] composed
of minibatch entries
t, ..., t[batch_size - 1] and tiles it to have a
[batch_size * multiplier, s0, s1, ...] composed of minibatch
t, t, ..., t, t, ... where each minibatch entry is
multiplier: Python int.
name: Name scope for any created operations.
A (possibly nested structure of)
[batch_size * multiplier, ...].
ValueError: if tensor(s)
tdo not have a statically known rank or the rank is < 1.