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.
SpatialMappingConversionStage Member List

This is the complete list of members for SpatialMappingConversionStage, including all inherited members.

__init__(self, list[StageCallable] list_of_callables, *Accelerator accelerator, LayerNode layer, **Any kwargs)SpatialMappingConversionStage
zigzag::stages::stage::Stage.__init__(self, list["StageCallable"] list_of_callables, **Any kwargs)Stage
__iter__(self)Stage
acceleratorSpatialMappingConversionStage
calc_unrolled_loop_size_on_early_oa_dims(self, OADimension oa_dim, LayerDim loop_dim_unrolled, SpatialMapping user_spatial_mapping)SpatialMappingConversionStage
check_if_oa_dim_mapping_is_first_max(self, OADimension oa_dim, LayerDim loop_dim_unrolled, SpatialMapping user_spatial_mapping)SpatialMappingConversionStage
convert_user_spatial_mapping(self, SpatialMapping user_spatial_mapping)SpatialMappingConversionStage
generate_limited_user_spatial_mapping(self, LayerDimSizes layer_dim_sizes, OADimension oa_dim, tuple[LayerDim, UnrollFactor] spatial_loop, SpatialMapping user_spatial_mapping, SpatialMapping limited_user_spatial_mapping, bool allow_decimal_sm_loop_size=True)SpatialMappingConversionStage
generate_mapping_per_mem_lvl(self, SpatialMapping user_spatial_mapping)SpatialMappingConversionStage
is_leaf(self)Stage
kwargsStage
layerSpatialMappingConversionStage
list_of_callablesStage
memory_operand_linksSpatialMappingConversionStage
run(self)SpatialMappingConversionStage
user_spatial_mappingSpatialMappingConversionStage