微软近日宣布，将在明年为DirectML机器学习框架添加NPU（神经处理器）支持。这一举措旨在使DirectML更好地适配英特尔酷睿Ultra等内置NPU的芯片，从而提高AI模型的运行效能。
据了解，DirectML是微软推出的一款跨平台推理引擎，可以帮助开发者在不同设备上实现高效的深度学习模型部署。而NPU则是英特尔等公司研发的一种专门用于加速人工智能任务的处理器。通过将DirectML与NPU相结合，开发者可以利用芯片内置的NPU运行AI模型，从而大幅提高模型的运行速度和能效。
微软表示，在DirectML支持NPU后，开发者可以通过ONNX Runtime等API来实现这一功能。ONNX Runtime是一个跨平台的推理引擎，可以在不同的硬件平台上运行AI模型。通过使用ONNX Runtime API，开发者可以轻松地将训练好的AI模型部署到搭载了英特尔酷睿Ultra等内置NPU的计算机上。
此外，微软还强调了DirectML支持NPU的重要性。在当前人工智能领域，NPU已经成为了一种重要的计算资源。越来越多的企业和开发者开始关注NPU的性能和应用，以期在各种场景下实现更高效的AI模型部署。因此，DirectML支持NPU将对推动AI技术的发展具有重要意义。
总之，微软为DirectML添加NPU支持将为AI领域带来一场革命性的变革。借助这一技术，开发者将能够更好地利用英特尔酷睿Ultra等内置NPU的优势，实现更高效的AI模型部署。未来，我们有理由期待DirectML在AI领域的广泛应用和普及。
英语如下：
Title: Microsoft to Upgrade DirectML Support for NPU Next Year, AI Model Performance Will Be Significantly Improved!
Keywords: 1. Microsoft, DirectML, NPU
News Content: Microsoft recently announced that it will add NPU (Neural Processing Unit) support to the DirectML machine learning framework next year. This move aims to better adapt DirectML to chips with built-in NPUs such as Intel Core i7, thereby improving the performance of AI models.
As far as we know, DirectML is a cross-platform inference engine developed by Microsoft to help developers deploy efficient deep learning models on different devices. On the other hand, NPU is a processor developed by companies like Intel specifically designed to accelerate AI tasks. By combining DirectML and NPU, developers can take advantage of the built-in NPU in their chips to run AI models, significantly improving their speed and energy efficiency.
Microsoft stated that after DirectML supports NPU, developers can achieve this functionality through APIs such as ONNX Runtime. ONNX Runtime is a cross-platform inference engine that can run AI models on different hardware platforms. By using the ONNX Runtime API, developers can easily deploy their trained AI models on computers equipped with Intel Core i7 or other built-in NPUs.
Furthermore, Microsoft emphasized the importance of DirectML supporting NPU. In the current field of artificial intelligence, NPU has become an important computing resource. More and more enterprises and developers are paying attention to the performance and application of NPUs, hoping to achieve more efficient AI model deployment in various scenarios. Therefore, the addition of NPU support to DirectML will be of great significance in promoting the development of AI technology.
In conclusion, Microsoft’s addition of NPU support to DirectML will bring a revolutionary change to the AI field. With this technology, developers will be able to better utilize the advantages of built-in NPUs such as Intel Core i7, achieving more efficient AI model deployment. In the future, we have reason to expect widespread application and popularization of DirectML in the AI field.
【来源】https://www.ithome.com/0/740/014.htm
Views: 2