This guide assumes:
- You use the TensorFlow or Caffe neural network framework.
- Your machine learning workflow is carried out using 32-bit floating-point data types.
- The Arm NN SDK build environment is configured for the required parser. For more information on how to do this configuration, see Configuring the Arm NN SDK build environment for TensorFlow and Configuring the Arm NN SDK build environment for Caffe.