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.
|
A class to represent a constraint on a permutation that requires a predefined order for some or all elements. More...
Public Member Functions | |
def | __init__ (self, dict[int, LayerDim] static_positions) |
bool | is_valid (self, list[Any] permutation) |
def | is_empty (self) |
Public Attributes | |
static_positions | |
A class to represent a constraint on a permutation that requires a predefined order for some or all elements.
def __init__ | ( | self, | |
dict[int, LayerDim] | static_positions | ||
) |
def is_empty | ( | self | ) |
Reimplemented from PermutationConstraint.
bool is_valid | ( | self, | |
list[Any] | permutation | ||
) |
static_positions |