Remote ML
I have no idea what I am doing wrong for remote ML
My remote ml file
it loads into memory, but GPU utilisation is 0%
services:
immich-machine-learning:
image: ghcr.io/immich-app/immich-machine-learning:release-cuda
container_name: immich-ml
restart: unless-stopped
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]
environment:
- NVIDIA_VISIBLE_DEVICES=all
- NVIDIA_DRIVER_CAPABILITIES=compute,utility
- TRANSFORMERS_CACHE=/cache
volumes:
- ./cache:/cache
ports:
- "3003:3003"services:
immich-machine-learning:
image: ghcr.io/immich-app/immich-machine-learning:release-cuda
container_name: immich-ml
restart: unless-stopped
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]
environment:
- NVIDIA_VISIBLE_DEVICES=all
- NVIDIA_DRIVER_CAPABILITIES=compute,utility
- TRANSFORMERS_CACHE=/cache
volumes:
- ./cache:/cache
ports:
- "3003:3003"My remote ml file
it loads into memory, but GPU utilisation is 0%
