concepts.dsl.tensor_state#

Data structures for tensor states. A tensor state is composed of tensor values. Intuitively, each state can be represented as a dictionary, mapping from feature names to tensors. Semantically, this dictionary represents the state of the system (e.g., a scene, a collection of robot and object states in an environment, etc.).

Functions

concat_states(*args)

Concatenate a list of states into a batch state.

Classes

MultidimensionalArrayInterface

A multi-dimensional array inferface.

NamedObjectStateMixin

A state type mixin with named objects.

NamedObjectTensorState

A state type with named objects.

StateObjectReference

The StateObjectReference class represents a reference to an object in the state.

TensorState

A state representation is essentially a mapping from feature names to tensors.

TensorStateBase

A state representation is essentially a mapping from feature names to tensors.

TensorValueDict

Basic tensor dict implementation.