HCLS Dev Summit: Accelerating Deep Learning Inference with TensorRT

Performance is a critical component of a successful deep learning application in health care. Biopharmaceutical innovation relies on high throughput and low latency to meet strict, and in some cases real-time, performance requirements while rapidly searching for signal in vast amounts of noise. Learn how computationally expensive deep learning models can be deployed efficiently using NVIDIA TensorRT for inference optimization and NVIDIA Triton Inference Server for deployment. We'll discuss the features and advantages of TensorRT and Triton, the different TensorRT workflows available to deep learning framework users, and finally how to deploy a TensorRT-optimized deep learning model through Triton.