Since the emergence of Deep Neural Networks (DNNs) as a prominent technique in the field of computer vision, the ImageNet classification challenge has played a major role in advancing the state-of-the-art. While accuracy figures have steadily increased, the resource utilization of winning models has not been properly taken into account. In this work, we present a comprehensive analysis of important metrics in practical applications: accuracy, memory footprint, parameters, operations count, inference time and power consumption. Key findings are: (1) fully connected layers are largely inefficient for smaller batches of images; (2) accuracy and inference time are in a hyperbolic relationship; (3) energy constraint are an upper bound on the maximum achievable accuracy and model complexity; (4) the number of operations is a reliable estimate of the inference time. We believe our analysis provides a compelling set of information that help design and engineer efficient DNNs.

本文对深度神经网络在计算机视觉领域中的重要应用指标进行全面分析，发现能耗与批次大小和架构无关，准确度和推断时间在双曲线关系中，能源约束是最大可实现准确度和模型复杂度的上限，操作次数是推断时间的可靠估计。

深度神经网络模型在实践应用中的分析