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.
CactiConfig Class Reference

Configuration for Cacti. More...

Public Member Functions

def __init__ (self)
 
def change_default_value (self, list[str] name_list, list[Any] new_value_list)
 
def write_config (self, list[Any] user_config, str path)
 
def call_cacti (self, str path)
 
def cacti_auto (self, list[Any] user_input, str path)
 user_input format can be 1 out of these 3: user_input = ['default'] user_input = ['single', [['mem_type', 'technology', ...], ['"ram"', 0.028, ...]] user_input = ['sweep', ['IO_bus_width'/'']] More...
 

Public Attributes

 baseline_config
 

Detailed Description

Configuration for Cacti.

Constructor & Destructor Documentation

◆ __init__()

def __init__ (   self)

Member Function Documentation

◆ cacti_auto()

def cacti_auto (   self,
list[Any]  user_input,
str  path 
)

user_input format can be 1 out of these 3: user_input = ['default'] user_input = ['single', [['mem_type', 'technology', ...], ['"ram"', 0.028, ...]] user_input = ['sweep', ['IO_bus_width'/'']]

Here is the call graph for this function:

◆ call_cacti()

def call_cacti (   self,
str  path 
)
Here is the caller graph for this function:

◆ change_default_value()

def change_default_value (   self,
list[str]  name_list,
list[Any]  new_value_list 
)

◆ write_config()

def write_config (   self,
list[Any]  user_config,
str  path 
)
Here is the caller graph for this function:

Member Data Documentation

◆ baseline_config

baseline_config

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