System Analysis/Design

 Architecture Overview of TensorFlow Lite

            TensorFlow  Lite support both Android and iOS platforms.  The initial step involves conversion of a trained TensorFlow model to TensorFlow Lite file format (.tflite) Using the TensorFlow Lite Converter. This converted model file is used in the application. Following are the important components for deploying the model as shown in the architecture diagram:

1. Java API: A wrapper around the C++ API (For Android)
2. C++ API: The C++ API is responsible for loading the model file and invoking the interpreter for further processing and execution. The same library is used for android and iOS.
3. Interpreter: Interpreter executes the model using the defined kernels. The interpreter supports selective kernel loading developers can also define their own custom kernel that can be used perform the execution. On few android devices, the interpreter uses the Android Neural Network API for hardware acceleration or default to using a CPU.     

Figure-1 : TensorFlow Architecture Lite

Figure-2: Process Of Flow Diagram