torch_concepts.EmbeddingVariable

class EmbeddingVariable(names: str | List[str], distribution=None, shape: int | Tuple | Size | List | None = None, dist_kwargs: dict | List[dict | None] | None = None, size: int | List[int] | None = None, members: List[str] | None = None)

A non-interpretable embedding variable.

May be observed, latent, or deterministic (via dist.Delta); the engine decides on a per-call basis whether the variable is observed.

__init__(names: str | List[str], distribution=None, shape: int | Tuple | Size | None = None, dist_kwargs: dict | List[dict | None] | None = None, size: int | List[int] | None = None, members: List[str] | None = None)

Methods

__init__(names[, distribution, shape, ...])
column_of(member)	Column span of member within this variable's event (last) dimension.
get_slice(labels)	Get slice or indices for concept(s) in the flattened tensor.
member(name)	A handle to a single member, usable as a parent (an edge to that member only).

Attributes

concept_slices	Precomputed mapping from concept name to slice in flattened tensor.
is_plate	Whether this variable was created with explicit named members.
param_sizes	Per-parameter output sizes for this variable's distribution.
plate	The plate this variable belongs to.
shape	Event shape as a torch.Size, e.g. torch.Size([4]) or torch.Size([3, 4]).
size	math.prod(self.shape).
variable_type	Short string tag identifying the variable kind.