quantization

Deploying efficient neural nets on mobiles is becoming increasingly important. This post explores the concept of quantized inference, and how it works in TensorFlow Lite.