Neural Network Inference on Moto 2022
7.5Baseline Latency (ms)Fine-Grained CPU-GPU Co-Execution
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Fine-Grained CPU-GPU Co-ExecutionNeural Network=ResNet-18, CPU Threads=32025.10 | 7.5 | 6.3 | 1.18 | 6.7 | 1.11 | |
| Fine-Grained CPU-GPU Co-ExecutionNeural Network=ResNet-34, CPU Threads=32025.10 | 14.7 | 12.2 | 1.2 | 12.9 | 1.14 | |
| Fine-Grained CPU-GPU Co-ExecutionNeural Network=VGG16, CPU Threads=32025.10 | 32 | 28.7 | 1.11 | 29.7 | 1.08 | |
| Fine-Grained CPU-GPU Co-ExecutionNeural Network=Inception-v3, CPU Threads=32025.10 | 52 | 38.8 | 1.34 | 41.1 | 1.27 |