feat(ml)!: cuda and openvino acceleration (#5619)

* cuda and openvino ep, refactor, update dockerfile

* updated workflow

* typing fixes

* added tests

* updated ml test gh action

* updated README

* updated docker-compose

* added compute to hwaccel.yml

* updated gh matrix

updated gh matrix

updated gh matrix

updated gh matrix

updated gh matrix

give up

* remove cuda/arm64 build

* add hwaccel image tags to docker-compose

* remove unnecessary quotes

* add suffix to git tag

* fixed kwargs in base model

* armnn ld_library_path

* update pyproject.toml

* add armnn workflow

* formatting

* consolidate hwaccel files, update docker compose

* update hw transcoding docs

* add ml hwaccel docs

* update dev and prod docker-compose

* added armnn prerequisite docs

* support 3.10

* updated docker-compose comments

* formatting

* test coverage

* don't set arena extend strategy for openvino

* working openvino

* formatting

* fix dockerfile

* added type annotation

* add wsl configuration for openvino

* updated lock file

* copy python3

* comment out extends section

* fix platforms

* simplify workflow suffix tagging

* simplify aio transcoding doc

* update docs and workflow for `hwaccel.yml` change

* revert docs

.github/workflows/docker.yml

@@ -25,15 +25,38 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - context: "machine-learning"
-            file: "machine-learning/Dockerfile"
-            image: "immich-machine-learning"
-            platforms: "linux/amd64,linux/arm64"
-          - context: "."
-            file: "server/Dockerfile"
-            image: "immich-server"
-            platforms: "linux/arm64,linux/amd64"
+          - image: immich-machine-learning
+            context: machine-learning
+            file: machine-learning/Dockerfile
+            platforms: linux/amd64,linux/arm64
+            device: cpu
 
+          - image: immich-machine-learning
+            context: machine-learning
+            file: machine-learning/Dockerfile
+            platforms: linux/amd64
+            device: cuda
+            suffix: -cuda
+
+          - image: immich-machine-learning
+            context: machine-learning
+            file: machine-learning/Dockerfile
+            platforms: linux/amd64
+            device: openvino
+            suffix: -openvino
+          
+          - image: immich-machine-learning
+            context: machine-learning
+            file: machine-learning/Dockerfile
+            platforms: linux/arm64
+            device: armnn
+            suffix: -armnn
+
+          - image: immich-server
+            context: .
+            file: server/Dockerfile
+            platforms: linux/amd64,linux/arm64
+            device: cpu
     steps:
       - name: Checkout
         uses: actions/checkout@v4
@@ -79,12 +102,12 @@ jobs:
             name=altran1502/${{matrix.image}},enable=${{ github.event_name == 'release' }}
           tags: |
             # Tag with branch name
-            type=ref,event=branch
+            type=ref,event=branch,suffix=${{ matrix.suffix }}
             # Tag with pr-number
-            type=ref,event=pr
+            type=ref,event=pr,suffix=${{ matrix.suffix }}
             # Tag with git tag on release
-            type=ref,event=tag
-            type=raw,value=release,enable=${{ github.event_name == 'release' }}
+            type=ref,event=tag,suffix=${{ matrix.suffix }}
+            type=raw,value=release,enable=${{ github.event_name == 'release' }},suffix=${{ matrix.suffix }}
 
       - name: Determine build cache output
         id: cache-target
@@ -106,5 +129,7 @@ jobs:
           push: ${{ !github.event.pull_request.head.repo.fork }}
           cache-from: type=registry,ref=ghcr.io/${{ github.repository_owner }}/immich-build-cache:${{matrix.image}}
           cache-to: ${{ steps.cache-target.outputs.cache-to }}
+          build-args: |
+            DEVICE=${{ matrix.device }}
           tags: ${{ steps.metadata.outputs.tags }}
           labels: ${{ steps.metadata.outputs.labels }}

.github/workflows/prepare-release.yml

@@ -83,5 +83,6 @@ jobs:
           files: |
             docker/docker-compose.yml
             docker/example.env
-            docker/hwaccel.yml
+            docker/hwaccel.ml.yml
+            docker/hwaccel.transcoding.yml
             *.apk

.github/workflows/test.yml

@@ -225,7 +225,7 @@ jobs:
           cache: "poetry"
       - name: Install dependencies
         run: |
-          poetry install --with dev
+          poetry install --with dev --with cpu
       - name: Lint with ruff
         run: |
           poetry run ruff check --output-format=github app export