Open issues for triton-inference-server (round 2) #17
Description
Tracking the second round of issues submitted to triton-inference-server:
- Control number of threads used by CPU server triton-inference-server/server#2018: Control number of threads used by CPU server
- Stricter model versioning triton-inference-server/server#2019: Stricter model versioning
- Ability to disable or redirect cout/cerr triton-inference-server/server#2020: Ability to disable or redirect cout/cerr
- Support input/output compression triton-inference-server/server#2021: Support input/output compression
- Ragged batching support for ML backends triton-inference-server/server#2158: Ragged batching support for ML backends
- Support different network protocols triton-inference-server/server#2159: Support different network protocols
- CMake problems for client library in v2.11.0 triton-inference-server/server#3141: CMake problems for client library in v2.11.0