Skip to main content
. 2023 Jan 19;13(3):385. doi: 10.3390/diagnostics13030385

Table 9.

Performance of CPU/TPU/GPU Comparisons of the proposed ShuffleNet-Light model.

Batch Size Epochs CPU/TPU/GPU (MS)
64 40 600/400/500
128 40 800/450/550
256 40 900/500/600
512 40 900/500/600
1024 40 900/500/600

MS: milliseconds, CPU: central processing unit, GPU: graphical processing unit, and TPU: Tensor Processing Units.