The Intel® Movidius™ Neural Compute SDK provides tools for profiling, tuning, and compiling a deep neural network (DNN) model on a development computer (host system).
|mvNCCompile||Converts a Caffe/TensorFlow* network and associated weights to an internal Intel® Movidius™ compiled format for use with the Intel® Movidius™ Neural Compute API.|
|mvNCProfile||Provides layer-by-layer statistics to evaluate the performance of Caffe/TensorFlow networks on your neural compute device.|
|mvNCCheck||Compares the inference results from running the network on your neural compute device vs. Caffe/TensorFlow for network compilation validation.|
See the documentation for each tool for more information.
See TensorFlow* SSD config for more information on using the NCSDK toolkit with a TensorFlow SSD network.