Opt-in capability interface: an InferenceEngineProvider whose models can
be open side by side, which FlutterEdgeAi.openModel requires.
An engine that implements this allows its createModel to run again while
models it built earlier are still open, and to be called again before an
earlier call has returned (it serialises what must be serialised), and
those models are independent: closing one leaves the others working. Each
holds its own weights and context, and the caller decides how many fit in
memory.
Kept separate from InferenceEngineProvider for the reason
HuggingFaceResolverSource is: engines implements the provider, so a new
member there would break every third-party engine. An engine opts in with
implements InferenceEngineProvider, MultiModelEngine.
Properties
- hashCode → int
-
The hash code for this object.
no setterinherited
- runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
Methods
-
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
toString(
) → String -
A string representation of this object.
inherited
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited