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.
Classes | Variables
zigzag.mapping.spatial_mapping Namespace Reference

Classes

class  MappingSingleOADim
 Spatial unrolling for a single OADimension. More...
 
class  SpatialMapping
 Spatial unrollings defined for every operational array dimension. More...
 
class  SpatialMappingHint
 Suggested LayerDims to be unrolled for every OADimension. More...
 

Variables

 logger = logging.getLogger(__name__)
 

Variable Documentation

◆ logger

logger = logging.getLogger(__name__)