Why Are Precision Values Different When Inferencing in GNA_SW_FP32 and GNA_HW Execution Modes?
Content Type: Product Information & Documentation | Article ID: 000089133 | Last Reviewed: 05/20/2022
When inferencing in GNA_SW_FP32 execution mode, parameters and calculations are substituted from low precision to floating point (FP32).
When inferencing in GNA_HW execution mode, the default precision value is I16.