Tile the batch dimension of a (possibly nested structure of) tensor(s).
tf_agents.utils.nest_utils.tile_batch(
tensors: tf_agents.typing.types.NestedTensor
,
multiplier: tf_agents.typing.types.Int
)
Copied from tensorflow/contrib/seq2seq/python/ops/beam_search_decoder.py
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[0], ..., t[batch_size - 1]
and tiles it to have a shape
[batch_size * multiplier, s0, s1, ...]
composed of minibatch entries
t[0], t[0], ..., t[1], t[1], ...
where each minibatch entry is repeated
multiplier
times.
Args |
tensors
|
A nested structure of Tensor shaped [batch_size, ...] .
|
multiplier
|
Python int or a Tensor. Note that if the multiplier is a tensor
the shape can not be ensured.
|
Returns |
A (possibly nested structure of) Tensor shaped
[batch_size * multiplier, ...] .
|
Raises |
ValueError
|
if tensor(s) t do not have a statically known rank or
the rank is < 1.
|