MCPcopy Create free account
hub / github.com/pytorch/executorch / MemoryAlgoResult

Class MemoryAlgoResult

exir/memory_planning.py:743–751  ·  view source on GitHub ↗

This is the result returned by a memory planning algorithm that is invoked by memory_planning_algorithm_suite. It contains the allocation decisions of that algorithm for all the specs, and the size of the buffer that was used for different memory hierarchies.

Source from the content-addressed store, hash-verified

741
742@dataclass
743class MemoryAlgoResult:
744 """This is the result returned by a memory planning algorithm that is
745 invoked by memory_planning_algorithm_suite. It contains the allocation
746 decisions of that algorithm for all the specs, and the size of the buffer
747 that was used for different memory hierarchies.
748 """
749
750 spec_dict: Dict[TensorSpec, SpecAllocResult]
751 bufsizes: List[int]
752
753
754def materialize_buffer(

Callers 2

greedyFunction · 0.85
naiveFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected