ZigZag - Deep Learning Hardware Design Space Exploration
This repository presents the novel version of our tried-and-tested hardware Architecture-Mapping Design Space Exploration (DSE) Framework for Deep Learning (DL) accelerators. ZigZag bridges the gap between algorithmic DL decisions and their acceleration cost on specialized accelerators through a fast and accurate hardware cost estimation.
|
Port allocation for a single memory instance. More...
Public Member Functions | |
def | __init__ (self, dict[MemoryOperand, dict[DataDirection, str]] data) |
dict[DataDirection, str] | get_alloc_for_mem_op (self, MemoryOperand mem_op) |
Public Attributes | |
data | |
Port allocation for a single memory instance.
Stores which ports are available for which memory operands and their corresponding direction.
def __init__ | ( | self, | |
dict[MemoryOperand, dict[DataDirection, str]] | data | ||
) |
dict[DataDirection, str] get_alloc_for_mem_op | ( | self, | |
MemoryOperand | mem_op | ||
) |
data |