Samples the given nest of specs.
tf_agents.specs.sample_spec_nest(
structure, seed=None, outer_dims=(), minimum=None, maximum=None
)
Args |
structure
|
A nest of TensorSpec .
|
seed
|
A seed used for sampling ops
|
outer_dims
|
An optional Tensor specifying outer dimensions to add to the
spec shape before sampling.
|
minimum
|
An optional numeric value. If set, numeric specs within the nest
(both bounded and unbounded) will be restricted to this minimum.
|
maximum
|
Similar to the above but with maximums.
|
Returns |
A nest of sampled values following the ArraySpec definition.
|
Raises |
TypeError
|
If spec is an unknown type.
|
NotImplementedError
|
If outer_dims is not statically known but nest
contains a SparseTensorSpec .
|