TensorBufferMode enum
Tensor buffer allocation mode for CompiledModel I/O.
managed uses LiteRT-managed buffers and copies host data through the
documented lock/write/unlock and lock/read/unlock path. hostMemory wraps
package-owned, 64-byte-aligned host memory with
LiteRtCreateTensorBufferFromHostMemory.
Host-memory buffers are opt-in because performance is model- and accelerator-dependent: they can reduce lock/copy overhead for some strict GPU models, but are slower for others. hostMemory is native-only; the web implementation supports managed buffers exclusively.
Values
- managed → const TensorBufferMode
- hostMemory → const TensorBufferMode
Properties
- hashCode → int
-
The hash code for this object.
no setterinherited
- index → int
-
A numeric identifier for the enumerated value.
no setterinherited
- name → String
-
Available on Enum, provided by the EnumName extension
The name of the enum value.no setter - 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
Constants
-
values
→ const List<
TensorBufferMode> - A constant List of the values in this enum, in order of their declaration.