immich/machine-learning/immich_ml/models
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
..
clip feat: ocr (#18836) 2025-10-27 14:09:55 +00:00
facial_recognition fix(ml): stabilize MIGraphX inference (#28444) 2026-05-26 18:41:56 +00:00
ocr fix(ml): batch size setting (#26524) 2026-03-05 12:01:47 -05:00
__init__.py feat: ocr (#18836) 2025-10-27 14:09:55 +00:00
base.py feat: ocr (#18836) 2025-10-27 14:09:55 +00:00
cache.py chore(ml): installable package (#17153) 2025-03-27 19:49:09 +00:00
constants.py feat: ROCm 7.2 and MIGraphX support (#26178) 2026-02-26 16:52:26 +00:00
transforms.py fix(deps): update machine-learning (#19803) 2025-08-11 18:07:49 -04:00