Skip to main content
. 2020 Feb 28;14:119. doi: 10.3389/fnins.2020.00119

Figure 6.

Figure 6

Inference performance variation due to (A) #Training-Timesteps and (B) #Inference-Timesteps. T# in (A) indicates number of time-steps used for training. (A) shows that inference accuracy starts to saturate as #training-timesteps increase. In (B), the zoomed version on the inset shows that the SNN trained with the proposed scheme performs very well even with only 30 time-steps while the peak performance occurs around 100 time-steps.