Digital Computer Modulation Signal Classification Based on Neural Network

Table 4

Time-consuming situation of fixed parameter model compression method.

Platform	Original model time ()	Compression model time ()	Reduction ratio
NVIDIA GTX1060	4.19	1.4665	0.65
NVIDIA RTX2060	2.114	0.3593	0.83
GT730M	8.601	2.5803	0.7
I3-8100	17.195	2.5792	0.85
NVIDIA RTX A4000	8.676	1.5616	0.82