Table 4 Baseline results for the NHP motor prediction task, for NHP Indy (96-channel data, top), and NHP Loco (192-channel data, bottom)
From: The neurobench framework for benchmarking neuromorphic computing algorithms and systems
Baseline | R2 | Footprint (bytes) | Model Exec. Rate (Hz) | Connection Sparsity | Activation Sparsity | SynOps (per model exec.) | ||
|---|---|---|---|---|---|---|---|---|
Dense | Eff_MACs | Eff_ACs | ||||||
ANN | 0.593 | 20824 | 250 | 0.0 | 0.683 | 4704 | 3836 | 0 |
0.558 | 33496 | 250 | 0.0 | 0.668 | 7776 | 6103 | 0 | |
SNN | 0.593 | 19648 | 250 | 0.0 | 0.997 | 4900 | 0 | 276 |
0.568 | 38848 | 250 | 0.0 | 0.999 | 9700 | 0 | 551 | |
