Example: bachelor of science
Search results with tag "Tensorrt"
8-bit Inference with TensorRT - NVIDIA
on-demand.gputechconf.comMay 8, 2017. Intro Goal: Convert FP32 CNNs into INT8 without significant accuracy loss. Why: INT8 math has higher throughput, and lower memory requirements. Challenge: INT8 has significantly lower precision and dynamic range than FP32.
S8495: DEPLOYING DEEP NEURAL NETWORKS AS-A …
on-demand.gputechconf.comPrethvi Kashinkunti, Solutions Architect Alec Gunny, Solutions Architect S8495: DEPLOYING DEEP NEURAL NETWORKS AS-A-SERVICE USING TENSORRT AND NVIDIA-DOCKER