Le projet THINK

Model Converter converts a public model into Inference Engine IR format using Model Optimizer

Near Memory Compute is becoming important for future AI processing systems that need improvement in system performance and energy-efficiency. The Von Neumann computing model requires data to commute from memory to compute and this data movement burns energy. Is it time for NMC to solve this data movement bottleneck? This blog addresses this question and is inspired by Intel Fellow, Dr. Frank Hady’s recent presentation at the International Solid State Circuits Conference (ISSCC), titled “We have rethought our commute; Can we rethink our data’s commute?” 

Intel Labs is developing an automated hardware-aware model optimization tool called BootstrapNAS to simplify the optimization of pre-trained AI models on Intel hardware, including Intel® Xeon® Scalable processors, which delivers built-in AI acceleration and flexibility. The tool will provide considerable time savings with respect to finding an optimal model design for a given AI platform, while simultaneously improving performance significantly.

Developers, like yourself, can now leverage model caching through the OpenVINO Execution Provider for ONNX Runtime, a product that accelerates inferencing of ONNX models using ONNX Runtime API’s while using OpenVINO™ toolkit as a backend. With the OpenVINO Execution Provider, ONNX Runtime delivers better inferencing performance on the same hardware compared to generic acceleration on Intel® CPU, GPU, and VPU.

Ce site contient les actualités du projet THINK ainsi que des liens utiles sur les techniques neuronales (cours, systèmes de développements, résultats comparaisons, etc …