fil_backend icon indicating copy to clipboard operation
fil_backend copied to clipboard

Published 20 hours ago • verified publisher icon triton-inference-server

Integrate experimental FIL in the FIL backend

Open hcho3 opened this issue 2 years ago • 0 comments

  • Use use_experimental_optimizations flag to selectively enable the new FIL
  • Enable the new FIL for both CPU and GPU inference workload. Note: requires https://github.com/rapidsai/cuml/pull/5559 to function.
  • Implement common postprocessors to process the output of the new FIL.

hcho3 avatar Aug 17 '23 05:08 hcho3