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.
exploit_data_locality_stages.py File Reference

Classes

class  SearchInterLayerDataLocalityStage
 Class for searching lowest allowed memory level per operand per layer. More...
 
class  ExploitInterLayerDataLocalityStage
 This stage must be processed behind WorkloadStage. More...
 

Namespaces

 zigzag.stages.exploit_data_locality_stages
 ZigZag simulates a workload layer-by-layer.
 

Variables

 logger = logging.getLogger(__name__)