Crate spaces

source ·
Expand description

Set/space primitives for defining machine learning problems.

spaces provides set/space primitives to be used for defining properties of machine learning problems. Traits such as Space, and it’s derivatives, may be used to define state/action spaces, for example. Mappings between different spaces may also be defined using traits such as Surjection to streamline many common preprocessing and type conversion tasks.

Modules

Continuous spaces module.
Discrete spaces module.
Normalisation utilities module.
Product spaces module.

Enums

Measure of the cardinality (#) of a set.

Traits

Trait for defining spaces bounded to lie on an interval I.
Trait for defining spaces containing a finite set of values.
Trait for defining geometric spaces.
Trait for types that implement a mapping from values of one set onto another.

Type Definitions

2d array type.
1d array type.