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 LomaEngine, including all inherited members.
__init__(self, *Accelerator accelerator, LayerNode layer, SpatialMappingInternal spatial_mapping, TemporalMappingType mapping_type, int|None loma_lpf_limit=None, **Any kwargs) | LomaEngine | |
accelerator | LomaEngine | |
compute_nb_permutations(self) | LomaEngine | |
constraints | LomaEngine | |
find_smallest_non_static_pf(self, LayerDim layer_dim) | LomaEngine | |
get_prime_factors(self) | LomaEngine | |
get_temporal_loops(self) | LomaEngine | |
has_constraints | LomaEngine | |
layer | LomaEngine | |
limit_lpfs(self) | LomaEngine | |
lpf_limit | LomaEngine | |
lpfs | LomaEngine | |
mapping_type | LomaEngine | |
memory_hierarchy | LomaEngine | |
nb_permutations | LomaEngine | |
ordering_generator(self) | LomaEngine | |
reduce_static_fps(self) | LomaEngine | |
run(self) | LomaEngine | |
set_constraints(self, list[PermutationConstraint] constraints) | LomaEngine | |
show_progress_bar | LomaEngine | |
spatial_mapping | LomaEngine | |
temporal_loop_dim_size | LomaEngine | |
temporal_loop_pf_count_sums | LomaEngine | |
temporal_loop_pf_counts | LomaEngine | |
update_min_lpf_factor(self, dict[LayerDim, UnrollFactor] loop_sizes) | LomaEngine |