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.
|
Class that passes through all results yielded by substages, but saves the results as a json list to a file at the end of the iteration. More...
Public Member Functions | |
def | __init__ (self, list[StageCallable] list_of_callables, *str dump_folder, **Any kwargs) |
def | run (self) |
Run the complete save stage by running the substage and saving the CostModelEvaluation json representation. More... | |
def | save_to_json (self, object obj, str filename) |
![]() | |
def | __init__ (self, list["StageCallable"] list_of_callables, **Any kwargs) |
def | __iter__ (self) |
bool | is_leaf (self) |
Public Attributes | |
dump_folder | |
![]() | |
kwargs | |
list_of_callables | |
Class that passes through all results yielded by substages, but saves the results as a json list to a file at the end of the iteration.
def __init__ | ( | self, | |
list[StageCallable] | list_of_callables, | ||
*str | dump_folder, | ||
**Any | kwargs | ||
) |
dump_folder | Output folder for dumps |
kwargs | any kwargs, passed on to substages |
def run | ( | self | ) |
Run the complete save stage by running the substage and saving the CostModelEvaluation json representation.
Reimplemented from Stage.
def save_to_json | ( | self, | |
object | obj, | ||
str | filename | ||
) |
dump_folder |