Method infer

samples/sampleIOFormats/sampleIOFormats.cpp:426–460

\brief Runs the TensorRT inference engine for this sample \details This function is the main execution function of the sample. It allocates the buffer, sets inputs, executes the engine, and verifies the output.

//!          the buffer, sets inputs, executes the engine, and verifies the output.
//!
bool SampleIOFormats::infer(SampleBuffer& inputBuf, SampleBuffer& outputBuf)
{
    auto const devInput = mallocCudaMem<uint8_t>(inputBuf.getBufferSize());
    auto devOutput = mallocCudaMem<uint8_t>(outputBuf.getBufferSize());

    CHECK(cudaMemcpy(devInput.get(), inputBuf.buffer, inputBuf.getBufferSize(), cudaMemcpyHostToDevice));

    auto context = SampleUniquePtr<nvinfer1::IExecutionContext>(mEngine->createExecutionContext());
    if (!context)
    {
        return false;
    }

    // Create CUDA stream for the execution of this inference.
    cudaStream_t stream;
    CHECK(cudaStreamCreate(&stream));

    void* bindings[2] = {devInput.get(), devOutput.get()};

    // Asynchronously enqueue the inference work
    if (!context->enqueueV2(bindings, stream, nullptr))
    {
        return false;
    }

    // Wait for the work in the stream to complete
    CHECK(cudaStreamSynchronize(stream));

    // Release stream
    CHECK(cudaStreamDestroy(stream));

    CHECK(cudaMemcpy(outputBuf.buffer, devOutput.get(), outputBuf.getBufferSize(), cudaMemcpyDeviceToHost));

    return true;
}

//!
//! \brief Reads the digit map from file