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.
|
Contains the size of each computation loop as defined in the workload, e.g. More...
Public Member Functions | |
def | __init__ (self, dict[LayerDim, UnrollFactor] data) |
list[LayerDim] | layer_dims (self) |
UnrollFactor | total_size (self) |
def | items (self) |
def | copy (self) |
def | __setitem__ (self, LayerDim key, UnrollFactor value) |
def | __delitem__ (self, LayerDim key) |
def | __add__ (self, "LayerDimSizes" other) |
![]() | |
def | __init__ (self, Any data) |
int | __len__ (self) |
Iterator[Any] | __iter__ (self) |
def | __getitem__ (self, Any key) |
bool | __contains__ (self, Any key) |
def | __str__ (self) |
def | __repr__ (self) |
Any | __jsonrepr__ (self) |
def | __eq__ (self, object other) |
def | __hash__ (self) |
Public Attributes | |
data | |
![]() | |
data | |
Contains the size of each computation loop as defined in the workload, e.g.
‘{'B’: 1, 'K': 32, 'C': 64, 'OY': 28, 'OX': 28, 'FY': 1, 'FX': 1, 'G': 1`
def __init__ | ( | self, | |
dict[LayerDim, UnrollFactor] | data | ||
) |
def __add__ | ( | self, | |
"LayerDimSizes" | other | ||
) |
def __delitem__ | ( | self, | |
LayerDim | key | ||
) |
def __setitem__ | ( | self, | |
LayerDim | key, | ||
UnrollFactor | value | ||
) |
def copy | ( | self | ) |
def items | ( | self | ) |
list[LayerDim] layer_dims | ( | self | ) |
UnrollFactor total_size | ( | self | ) |
data |