mirror-immich/machine-learning/immich_ml
Fabian Wimberger 53a24783f5
fix(ml): stabilize MIGraphX inference (#28444)
* fix: stabilize ROCm MIGraphX inference

Serialize MIGraphX session runs so lazy compiles cannot overlap within a worker.

Use a fixed face-recognition batch size for MIGraphX to avoid compiling a new program for each detected face count.

* fix(ml): increase ROCm worker timeout

* fix(ml): narrow MIGraphX compile locking

* docs: format environment variables table

* docs: apply prettier to environment variables table
2026-05-26 18:41:56 +00:00
..
models fix(ml): stabilize MIGraphX inference (#28444) 2026-05-26 18:41:56 +00:00
sessions fix(ml): stabilize MIGraphX inference (#28444) 2026-05-26 18:41:56 +00:00
__init__.py chore(ml): installable package (#17153) 2025-03-27 19:49:09 +00:00
__main__.py fix(ml): ipv6 check (#22735) 2025-10-07 12:24:23 -04:00
config.py fix(ml): stabilize MIGraphX inference (#28444) 2026-05-26 18:41:56 +00:00
gunicorn_conf.py chore(ml): installable package (#17153) 2025-03-27 19:49:09 +00:00
log_conf.json chore(ml): installable package (#17153) 2025-03-27 19:49:09 +00:00
main.py chore(ml)!: remove deprecated envs (#28326) 2026-05-09 22:40:05 +00:00
schemas.py feat(ml): add preload and fp16 settings for ocr (#23576) 2025-11-06 17:55:11 +00:00