Upgrade to Cuda12 and latest versions

.dockerignore

@@ -0,0 +1,10 @@
+.github
+.pytest_cache
+.ruff_cache
+.tox
+.venv
+.gitignore
+makefile
+__pycache__
+tests
+.vscode

.github/workflows/build-container.yaml

@@ -19,7 +19,8 @@ jobs:
     uses: ./.github/workflows/docker-build-action.yaml
     with:
       image: inference-pytorch-cpu
-      dockerfile: dockerfiles/pytorch/cpu/Dockerfile
+      dockerfile: dockerfiles/pytorch/Dockerfile
+      build_args: "BASE_IMAGE=ubuntu:22.04"
     secrets:
       TAILSCALE_AUTHKEY: ${{ secrets.TAILSCALE_AUTHKEY }}
       REGISTRY_USERNAME: ${{ secrets.REGISTRY_USERNAME }}
@@ -28,7 +29,7 @@ jobs:
     uses: ./.github/workflows/docker-build-action.yaml
     with:
       image: inference-pytorch-gpu
-      dockerfile: dockerfiles/pytorch/gpu/Dockerfile
+      dockerfile: dockerfiles/pytorch/Dockerfile
     secrets:
       TAILSCALE_AUTHKEY: ${{ secrets.TAILSCALE_AUTHKEY }}
       REGISTRY_USERNAME: ${{ secrets.REGISTRY_USERNAME }}

.github/workflows/integration-test-action.yaml

@@ -0,0 +1,69 @@
+on:
+  workflow_call:
+    inputs:
+      region:
+        type: string
+        required: false
+        default: "us-east-1"
+      hf_home:
+        required: false
+        type: string
+        default: "/mnt/hf_cache/"
+      hf_hub_cache:
+        required: false
+        type: string
+        default: "/mnt/hf_cache/hub"
+      run_slow:
+        required: false
+        type: string
+        default: "True"
+      test_path:
+        type: string
+        required: true
+      test_parallelism:
+        type: string
+        required: false
+        default: "4"
+      build_img_cmd:
+        type: string
+        required: false
+        default: "make inference-pytorch-gpu"
+      log_level:
+        type: string
+        required: false
+        default: "ERROR"
+      log_format:
+        type: string
+        required: false
+        default: "%(asctime)s %(levelname)s %(module)s:%(lineno)d %(message)s"
+      runs_on:
+        type: string
+        required: false
+        default: '["single-gpu", "nvidia-gpu", "t4", "ci"]'
+
+jobs:
+  pytorch-integration-tests:
+    runs-on: ${{ fromJson(inputs.runs_on) }}
+    env:
+      AWS_REGION: ${{ inputs.region }}
+      HF_HOME: ${{ inputs.hf_home }}
+      HF_HUB_CACHE: ${{ inputs.hf_hub_cache }}
+      RUN_SLOW: ${{ inputs.run_slow }}
+    steps:
+    - uses: actions/checkout@v4.1.1
+    - name: Docker Setup Buildx
+      uses: docker/setup-buildx-action@v3.0.0
+    - name: Docker Build
+      run: ${{ inputs.build_img_cmd }}
+    - name: Set up Python 3.11
+      uses: actions/setup-python@v2
+      with:
+        python-version: 3.11
+    - name: Install dependencies
+      run: pip install ".[torch, test]"
+    - name: Run local integration tests
+      run: |
+        python -m pytest \
+        ${{ inputs.test_path }} -n ${{ inputs.test_parallelism }} \
+        --log-cli-level='${{ inputs.log_level }}' \
+        --log-format='${{ inputs.log_format }}'

.github/workflows/integration-test.yaml

@@ -0,0 +1,44 @@
+name: Run Integration Tests
+
+on:
+  push:
+    paths-ignore:
+      - 'README.md'
+      - '.github/workflows/unit-test.yaml'
+      - '.github/workflows/quality.yaml'
+    branches:
+     - main
+  pull_request:
+  workflow_dispatch:
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
+  cancel-in-progress: true
+
+jobs:
+  pytorch-integration-local-gpu:
+    name: Local Integration Tests - GPU
+    uses: ./.github/workflows/integration-test-action.yaml
+    with:
+      test_path: "tests/integ/test_pytorch_local_gpu.py"
+      build_img_cmd: "make inference-pytorch-gpu"
+  pytorch-integration-remote-gpu:
+    name: Remote Integration Tests - GPU
+    uses: ./.github/workflows/integration-test-action.yaml
+    with:
+      test_path: "tests/integ/test_pytorch_remote_gpu.py"
+      build_img_cmd: "make inference-pytorch-gpu"
+  pytorch-integration-remote-cpu:
+    name: Remote Integration Tests - CPU
+    uses: ./.github/workflows/integration-test-action.yaml
+    with:
+      test_path: "tests/integ/test_pytorch_remote_cpu.py"
+      build_img_cmd: "make inference-pytorch-cpu"
+      runs_on: "['ci']"
+  pytorch-integration-local-cpu:
+    name: Local Integration Tests - CPU
+    uses: ./.github/workflows/integration-test-action.yaml
+    with:
+      test_path: "tests/integ/test_pytorch_local_cpu.py"
+      build_img_cmd: "make inference-pytorch-cpu"
+      runs_on: "['ci']"

.github/workflows/quality.yaml

@@ -2,6 +2,8 @@ name: Quality Check
 
 on:
   push:
+    paths-ignore:
+      - 'README.md'
     branches:
       - main
   pull_request:
@@ -16,10 +18,10 @@ jobs:
     runs-on: ubuntu-latest
     steps:
     - uses: actions/checkout@v2
-    - name: Set up Python 3.9
+    - name: Set up Python 3.11
       uses: actions/setup-python@v2
       with:
-        python-version: 3.9
+        python-version: 3.11
     - name: Install Python dependencies
       run: pip install -e .[quality]
     - name: Run Quality check