TensorFlow Lite is the TensorFlow deep learning framework for inference on edge devices. Similar to OpenVINO, TensorFlow Lite has built-in pre-trained deep learning modules. Alternatively, an existing model can be converted into TensorFlow Lite format for on-device inference. Currently, TensorFlow Lite provides inference support for PCs with a built-in or external camera, Android devices, iOS devices, Raspberry Pis, and tiny microcontrollers. Visit https://www.tensorflow.org/lite for details on TensorFlow Lite.
The TensorFlow Lite converter takes a TensorFlow model and generates a FlatBuffer tflite file. A FlatBuffer file is an efficient cross-platform library that can be used to access binary serialized data without the need for parsing. Serialized data is usually a text string. Binary serialized data is binary data written in string format. For...
