This application runs an inference on an image using GoogLeNet. Although the sample uses GoogLeNet as the default network, other classifier models can also be used (see Options section). The provided Makefile does the following
- Downloads the Caffe prototxt file and makes few changes necessary to work with the Intel® Neural Compute Stick (NCS1/NCS2) and the OpenVINO toolkit (tested with version 2020.1.
- Downloads and generates the required ilsvrc12 data.
- Downloads the caffemodel weights file from the Open Model Zoo.
- Compiles the model to an IR (Intermediate Representation) format file using the Model Optimizer. An IR is a static representation of the model that is compatitible with the OpenVINO Inference Engine API.
- There is a run.py provided that does a single inference on a provided image as an example on how to use the network using the OpenVINO Inference Engine Python API.
To run the example code do the following :
- Open a terminal and change directory to the sample base directory
- Type the following command in the terminal:
All development and testing has been done on Ubuntu 16.04 on an x86-64 machine.
This program requires:
- 1 Intel NCS device
- OpenVINO 2020.1 toolkit
Provided Makefile has various targets that help with the above mentioned tasks.
Runs the sample application.
Shows available targets.
Builds and/or gathers all the required files needed to run the application.
Gathers all of the required data need to run the sample.
Builds all of the dependencies needed to run the sample.
Compiles an IR file from a default model to be used when running the sample.
Checks required packages that aren’t installed as part of the OpenVINO installation.
Uninstalls requirements that were installed by the sample program.
Removes all the temporary files that are created by the Makefile.
|–ir IR_FILE||Absolute path to the neural network IR file. The default is: ../../caffe/SqueezeNet/squeezenet_v1.0.xml.|
|-l LABEL_FILE, –labels LABEL_FILE||Absolute path to labels file. The default is: ../../data/ilsvrc12/synset_words.txt.|
|-m NUMPY_MEAN_FILE, –mean NUMPY_MEAN_FILE||Network Numpy mean file. The default is: None.|
|-i IMAGE_FILE, –image IMAGE_FILE||Absolute path to the image file. The default is ../../data/images/nps_electric_guitar.png.|
|-n NUMBER_OF_RESULTS_TO_DISPLAY, –number_top NUMBER_OF_RESULTS_TO_DISPLAY||The number of inference results to display. The default is 1.|