Deploying Deep Neural Networks as a Service Using TensorRT and NVIDIA-Docker
Date:
Demo on converting trained recurrent and convolutional models from multiple deep learning libraries into accelerated inference runtime engines and exposing them via an inference service