latel Posts - Answer Overflow

latel

Posts Comments

IImmich

•Created by latel on 1/23/2025 in #help-desk-support

Error Loading image, but why

21 replies

IImmich

•Created by latel on 1/22/2025 in #help-desk-support

Does my machine support machine learning using openvino?

I see docs said

https://v1.124.2.archive.immich.app/docs/features/ml-hardware-acceleration#openvino The server must have a discrete GPU, i.e. Iris Xe or Arc. Expect issues when attempting to use integrated graphics.

But as far as I know, Iris XE is not a discrete GPU, it's integrated together in intel CPUs. How can I know my immich instance on "11th Gen Intel® Core™ i7-1165G7 @ 2.80GHz" is processing machine learning with openvino. here is my docker-compose settings

#
# WARNING: Make sure to use the docker-compose.yml of the current release:
#
# https://github.com/immich-app/immich/releases/latest/download/docker-compose.yml
#
# The compose file on main may not be compatible with the latest release.
#

name: immich

services:
  immich-machine-learning:
    container_name: immich_machine_learning
    # For hardware acceleration, add one of -[armnn, cuda, openvino] to the image tag.
    # Example tag: ${IMMICH_VERSION:-release}-cuda
    image: ghcr.io/immich-app/immich-machine-learning:${IMMICH_VERSION:-release}
    # extends: # uncomment this section for hardware acceleration - see https://immich.app/docs/features/ml-hardware-acceleration
    #   file: hwaccel.ml.yml
    #   service: cpu # set to one of [armnn, cuda, openvino, openvino-wsl] for accelerated inference - use the `-wsl` version for WSL2 where applicable
    volumes:
      - /mnt/user/tempo/immich:/cache
      - /dev/bus/usb:/dev/bus/usb
    env_file:
      - .env
    device_cgroup_rules:
      - 'c 189:* rmw'
    devices:
      - /dev/dri:/dev/dri
    restart: always
    healthcheck:
      disable: false

#
# WARNING: Make sure to use the docker-compose.yml of the current release:
#
# https://github.com/immich-app/immich/releases/latest/download/docker-compose.yml
#
# The compose file on main may not be compatible with the latest release.
#

name: immich

services:
  immich-machine-learning:
    container_name: immich_machine_learning
    # For hardware acceleration, add one of -[armnn, cuda, openvino] to the image tag.
    # Example tag: ${IMMICH_VERSION:-release}-cuda
    image: ghcr.io/immich-app/immich-machine-learning:${IMMICH_VERSION:-release}
    # extends: # uncomment this section for hardware acceleration - see https://immich.app/docs/features/ml-hardware-acceleration
    #   file: hwaccel.ml.yml
    #   service: cpu # set to one of [armnn, cuda, openvino, openvino-wsl] for accelerated inference - use the `-wsl` version for WSL2 where applicable
    volumes:
      - /mnt/user/tempo/immich:/cache
      - /dev/bus/usb:/dev/bus/usb
    env_file:
      - .env
    device_cgroup_rules:
      - 'c 189:* rmw'
    devices:
      - /dev/dri:/dev/dri
    restart: always
    healthcheck:
      disable: false

10 replies

IImmich

•Created by latel on 1/21/2025 in #help-desk-support

Why image thumbnail generation is not using hardware acceleration?

I see ffmpeg commands in immich containers

/usr/bin/ffmpeg -skip_frame nointra -sws_flags accurate_rnd+full_chroma_int -i /data/external/iCloud/someone/SharedSync-FABD9C9D-15B1-4FF2-835E-XXX/IMG_0558.MOV -y -fps_mode vfr -frames:v 1 -update 1 -v verbose -vf fps=12:eof_action=pass:round=down,thumbnail=12,select=gt(scene\,0.1)-eq(prev_selected_n\,n)+isnan(prev_selected_n)+gt(n\,20),trim=end_frame=2,reverse,scale=250:-2:flags=lanczos+accurate_rnd+full_chroma_int:out_range=pc,tonemapx=tonemap=hable:desat=0:p=bt709:t=bt709:m=bt709:r=pc:peak=100:format=yuv420p upload/thumbs/9acfe021-f0dc-445d-ba2c-5f8b9d07de7c/b3/93/b3937cb9-bf33-4a12-b1eb-29c5e828d195-thumbnail.webp

/usr/bin/ffmpeg -skip_frame nointra -sws_flags accurate_rnd+full_chroma_int -i /data/external/iCloud/someone/SharedSync-FABD9C9D-15B1-4FF2-835E-XXX/IMG_0558.MOV -y -fps_mode vfr -frames:v 1 -update 1 -v verbose -vf fps=12:eof_action=pass:round=down,thumbnail=12,select=gt(scene\,0.1)-eq(prev_selected_n\,n)+isnan(prev_selected_n)+gt(n\,20),trim=end_frame=2,reverse,scale=250:-2:flags=lanczos+accurate_rnd+full_chroma_int:out_range=pc,tonemapx=tonemap=hable:desat=0:p=bt709:t=bt709:m=bt709:r=pc:peak=100:format=yuv420p upload/thumbs/9acfe021-f0dc-445d-ba2c-5f8b9d07de7c/b3/93/b3937cb9-bf33-4a12-b1eb-29c5e828d195-thumbnail.webp

Which seems not using the hardware acceleration, I've tested video decoding/encoding using qsv_hevc is working ok in my immich server container.

6 replies

Gaming

Programming