TFLITE_GPU_INFERENCE_PREFERENCE_BALANCED constant

int const TFLITE_GPU_INFERENCE_PREFERENCE_BALANCED

Balance init latency and throughput. This option will result in slightly higher init latency than FAST_SINGLE_ANSWER but should have inference latency closer to SUSTAINED_SPEED.

Implementation

static const int TFLITE_GPU_INFERENCE_PREFERENCE_BALANCED = 2;