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.
|
This is the complete list of members for SalsaStage, including all inherited members.
__init__(self, list[StageCallable] list_of_callables, *Accelerator accelerator, LayerNode layer, SpatialMappingInternal spatial_mapping, TemporalMappingType temporal_mapping_type, **Any kwargs) | SalsaStage | |
zigzag::stages::stage::Stage.__init__(self, list["StageCallable"] list_of_callables, **Any kwargs) | Stage | |
__iter__(self) | Stage | |
best_cme | SalsaStage | |
cme_queue | SalsaStage | |
compare_cme_energy(self, CostModelEvaluation cme) | SalsaStage | |
compare_cme_latency(self, CostModelEvaluation cme) | SalsaStage | |
compare_stage | SalsaStage | |
engine | SalsaStage | |
is_leaf(self) | Stage | |
kwargs | Stage | |
list_of_callables | Stage | |
mapping_type | SalsaStage | |
number_of_core | SalsaStage | |
number_of_core_allocated | SalsaStage | |
opt_criterion_name | SalsaStage | |
run(self) | SalsaStage | |
spatial_mapping | SalsaStage | |
worker_list | SalsaStage |