- Generated by
1.9.1
|
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.
|
State of SALSA, storing an ordering, his temporal mapping and his energy value. More...
Public Member Functions | |
| def | __init__ (self, Accelerator accelerator, LayerNode layer, SpatialMappingInternal spatial_mapping, list[tuple[LayerDim, UnrollFactorInt]] ordering, str opt_criterion_name, TemporalMappingType mapping_type) |
| "SalsaState" | swap (self, int i, int j) |
| Swap between the element at position i and j in the ordering and return the new resulting state. More... | |
Public Attributes | |
| ordering | |
| accelerator | |
| layer | |
| spatial_mapping | |
| memory_hierarchy | |
| opt_criterion_name | |
| mapping_type | |
| temporal_mapping | |
| cme | |
| opt_criterion | |
State of SALSA, storing an ordering, his temporal mapping and his energy value.
| def __init__ | ( | self, | |
| Accelerator | accelerator, | ||
| LayerNode | layer, | ||
| SpatialMappingInternal | spatial_mapping, | ||
| list[tuple[LayerDim, UnrollFactorInt]] | ordering, | ||
| str | opt_criterion_name, | ||
| TemporalMappingType | mapping_type | ||
| ) |
| "SalsaState" swap | ( | self, | |
| int | i, | ||
| int | j | ||
| ) |
Swap between the element at position i and j in the ordering and return the new resulting state.
| accelerator |
| cme |
| layer |
| mapping_type |
| memory_hierarchy |
| opt_criterion |
| opt_criterion_name |
| ordering |
| spatial_mapping |
| temporal_mapping |
1.9.1