allowXnnpackAcceleration property
Enables/disables XNNPACK acceleration for TensorFlow ML models, which provides highly optimized implementations of floating-point neural network operators (Android only).
Default is true
Implementation
bool allowXnnpackAcceleration;