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.workload.layer_node Namespace Reference

Classes

class  LoopRelevancyInfo
 Per LayerOperand, store the Relevant, Irrelevant LayerDims, and which LayerDims are Partially Relevant to each other. More...
 
class  LayerNodeAttributes
 Packs the attributes needed to initialize a LayerNode. More...
 
class  MappingAttributes
 Packs the attributes needed to initialize a LayerNode, related to the mapping. More...
 
class  LayerNode
 Represents a single layer in a workload. More...
 

Variables

 logger = logging.getLogger(__name__)
 

Variable Documentation

◆ logger

logger = logging.getLogger(__name__)