tests/OnnxMobileNet-Armnn/OnnxMobileNet-Armnn.cpp

//
// Copyright © 2017 Arm Ltd. All rights reserved.
// SPDX-License-Identifier: MIT
//
#include "../InferenceTest.hpp"
#include "../ImagePreprocessor.hpp"
#include "armnnOnnxParser/IOnnxParser.hpp"

int main(int argc, char* argv[])
{
    int retVal = EXIT_FAILURE;
    try
    {
        // Coverity fix: The following code may throw an exception of type std::length_error.
        std::vector<ImageSet> imageSet =
        {
            {"Dog.jpg", 208},
            {"Cat.jpg", 281},
            {"shark.jpg", 2},
        };

        armnn::TensorShape inputTensorShape({ 1, 3, 224, 224 });

        using DataType = float;
        using DatabaseType = ImagePreprocessor<float>;
        using ParserType = armnnOnnxParser::IOnnxParser;
        using ModelType = InferenceModel<ParserType, DataType>;

        // Coverity fix: ClassifierInferenceTestMain() may throw uncaught exceptions.
        retVal = armnn::test::ClassifierInferenceTestMain<DatabaseType, ParserType>(
                     argc, argv,
                     "mobilenetv2-1.0.onnx", // model name
                     true,                           // model is binary
                     "data", "mobilenetv20_output_flatten0_reshape0", // input and output tensor names
                     { 0, 1, 2 },                    // test images to test with as above
                     [&imageSet](const char* dataDir, const ModelType&) {
                         // This creates create a 1, 3, 224, 224 normalized input with mean and stddev to pass to Armnn
                         return DatabaseType(
                             dataDir,
                             224,
                             224,
                             imageSet,
                             1.0,                             // scale
                             0,                               // offset
                             {{0.485f, 0.456f, 0.406f}},      // mean
                             {{0.229f, 0.224f, 0.225f}},      // stddev
                             DatabaseType::DataFormat::NCHW); // format
                     },
                     &inputTensorShape);
    }
    catch (const std::exception& e)
    {
        // Coverity fix: BOOST_LOG_TRIVIAL (typically used to report errors) may throw an
        // exception of type std::length_error.
        // Using stderr instead in this context as there is no point in nesting try-catch blocks here.
        std::cerr << "WARNING: OnnxMobileNet-Armnn: An error has occurred when running "
                     "the classifier inference tests: " << e.what() << std::endl;
    }
    return retVal;
}