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.
LayerDimRelation Class Reference

For the operand dimension that is not directly a loop dimension, a relation equations between them(operand dimension) and the loop dimension is required. More...

Inheritance diagram for LayerDimRelation:
Collaboration diagram for LayerDimRelation:

Public Member Functions

def __init__ (self, LayerDim dim_1, LayerDim dim_2, LayerDim dim_3, int coef_2, int coef_3)
 
- Public Member Functions inherited from LayerAttribute
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)
 

Static Public Member Functions

tuple[PrLoop, LoopList, PrScalingFactors] extract_pr_loop_info (list["LayerDimRelation"] relations)
 

Public Attributes

 dim_1
 
 dim_2
 
 dim_3
 
 coef_2
 
 coef_3
 
 data
 
- Public Attributes inherited from LayerAttribute
 data
 

Detailed Description

For the operand dimension that is not directly a loop dimension, a relation equations between them(operand dimension) and the loop dimension is required.

e.g. dim1 = coef2 * dim2 + coef3 * dim3

Constructor & Destructor Documentation

◆ __init__()

def __init__ (   self,
LayerDim  dim_1,
LayerDim  dim_2,
LayerDim  dim_3,
int  coef_2,
int  coef_3 
)

Member Function Documentation

◆ extract_pr_loop_info()

tuple[PrLoop, LoopList, PrScalingFactors] extract_pr_loop_info ( list["LayerDimRelation"]  relations)
static

TODO requires cleanup and documentation

Member Data Documentation

◆ coef_2

coef_2

◆ coef_3

coef_3

◆ data

data

◆ dim_1

dim_1

◆ dim_2

dim_2

◆ dim_3

dim_3

The documentation for this class was generated from the following file: