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.
architecture Directory Reference

Files

file  __init__.py
 
file  accelerator.py
 
file  get_cacti_cost.py
 
file  imc_array.py
 
file  imc_unit.py
 
file  memory_hierarchy.py
 
file  memory_instance.py
 
file  memory_level.py
 
file  memory_port.py
 
file  operational_array.py
 
file  operational_unit.py