Merge branch 'main' into tylertitsworth/open-reqs

Author: Tyler Titsworth

.github/workflows/container-ci.yaml

@@ -155,7 +155,7 @@ jobs:
     - name: Cleanup
       if: always()
       run: docker rmi -f ${{ secrets.REGISTRY }}/${{ secrets.REPO }}:${{ matrix.container }}
-    - uses: github/codeql-action/upload-sarif@429e1977040da7a23b6822b13c129cd1ba93dbb2 # v3.26.2
+    - uses: github/codeql-action/upload-sarif@2c779ab0d087cd7fe7b826087247c2c81f27bfa6 # v3.26.5
       with:
         sarif_file: '${{ matrix.container }}-scan.sarif'
         category: '${{ matrix.container }}'

.github/workflows/scorecard.yaml

@@ -53,6 +53,6 @@ jobs:
         name: SARIF file
         path: results.sarif
         retention-days: 5
-    - uses: github/codeql-action/upload-sarif@429e1977040da7a23b6822b13c129cd1ba93dbb2 # v3.26.2
+    - uses: github/codeql-action/upload-sarif@2c779ab0d087cd7fe7b826087247c2c81f27bfa6 # v3.26.5
       with:
         sarif_file: results.sarif

CONTRIBUTING.md

@@ -1,6 +1,6 @@
 # Contributing
 
-Thank you for considering contributing to Intel® AI Containers! We welcome your help to make this project better. Contributing to an open source project can be a daunting task, but the Intel AI Containers team is here to help you through the process. If at any point in this process you feel out of your depth or confused by our processes, please don't hesitate to reach out to a maintainer or file an [issue](https://github.com/intel/ai-containers/issues).
+Thank you for considering contributing to AI Containers! We welcome your help to make this project better. Contributing to an open source project can be a daunting task, but the Intel AI Containers team is here to help you through the process. If at any point in this process you feel out of your depth or confused by our processes, please don't hesitate to reach out to a maintainer or file an [issue](https://github.com/intel/ai-containers/issues).
 
 ## Getting Started
 
@@ -138,4 +138,4 @@ commit automatically with `git commit -s`.
 
 ## License
 
-Intel® AI Containers is licensed under the terms in [LICENSE](./LICENSE). By contributing to the project, you agree to the license and copyright terms therein and release your contribution under these terms.
+AI Containers is licensed under the terms in [LICENSE](./LICENSE). By contributing to the project, you agree to the license and copyright terms therein and release your contribution under these terms.

README.md

@@ -1,4 +1,4 @@
-# Intel® AI Containers
+# AI Containers
 
 [![OpenSSF Best Practices](https://www.bestpractices.dev/projects/8270/badge)](https://www.bestpractices.dev/projects/8270)
 [![OpenSSF Scorecard](https://api.securityscorecards.dev/projects/github.com/intel/ai-containers/badge)](https://securityscorecards.dev/viewer/?uri=github.com/intel/ai-containers)
@@ -28,7 +28,7 @@ docker login $REGISTRY
 docker pull $REGISTRY/$REPO:latest
 ```
 
-The maintainers of Intel® AI Containers use Azure to store containers, but an open source container registry like [harbor](https://github.com/goharbor/harbor) is preferred.
+The maintainers of AI Containers use Azure to store containers, but an open source container registry like [harbor](https://github.com/goharbor/harbor) is preferred.
 
 > [!WARNING]
 > You can optionally skip this step and use some placeholder values, however some container groups depend on other images and will pull from a registry that you have not defined and result in an error.

classical-ml/README.md

@@ -63,7 +63,7 @@ The images below additionally include [Jupyter Notebook](https://jupyter.org/) s
 
 ## Build from Source
 
-To build the images from source, clone the [Intel® AI Containers](https://github.com/intel/ai-containers) repository, follow the main `README.md` file to setup your environment, and run the following command:
+To build the images from source, clone the [AI Containers](https://github.com/intel/ai-containers) repository, follow the main `README.md` file to setup your environment, and run the following command:
 
 ```bash
 cd classical-ml

docs/requirements.txt

@@ -1,7 +1,7 @@
 mkdocs-callouts>=1.13.2
 mkdocs-git-authors-plugin>=0.8.0
 mkdocs-git-revision-date-localized-plugin>=1.2.5
-mkdocs-material==9.5.32
+mkdocs-material==9.5.33
 mkdocs-table-reader-plugin>=2.1.0
 mkdocs==1.6.0
 pandas>=2.0.3

mkdocs.yml

@@ -51,7 +51,7 @@ plugins:
     - read_csv
 repo_name: intel/ai-containers
 repo_url: https://github.com/intel/ai-containers
-site_name: Intel® AI Containers
+site_name: AI Containers
 #TODO: Get previous container versions in an easy way
 # https://squidfunk.github.io/mkdocs-material/setup/setting-up-versioning/
 theme:

python/README.md

@@ -15,7 +15,7 @@ The images below include variations for only the core packages in the [Intel® D
 
 ## Build from Source
 
-To build the images from source, clone the [Intel® AI Containers](https://github.com/intel/ai-containers) repository, follow the main `README.md` file to setup your environment, and run the following command:
+To build the images from source, clone the [AI Containers](https://github.com/intel/ai-containers) repository, follow the main `README.md` file to setup your environment, and run the following command:
 
 ```bash
 cd python

pytorch/Dockerfile

@@ -34,6 +34,7 @@ ARG BASE_IMAGE_TAG
 ARG PACKAGE_OPTION=pip
 ARG PYTHON_VERSION
 ARG PYTHON_BASE=${REGISTRY}/${REPO}:b-${GITHUB_RUN_NUMBER}-${BASE_IMAGE_NAME}-${BASE_IMAGE_TAG}-${PACKAGE_OPTION}-py${PYTHON_VERSION}-base
+ARG TORCHSERVE_BASE=${PYTHON_BASE}
 FROM ${PYTHON_BASE} AS ipex-base-pip
 
 WORKDIR /
@@ -181,13 +182,17 @@ RUN apt-get update && \
     intel-oneapi-runtime-mkl=${MKL_VER} \
     intel-oneapi-runtime-ccl=${CCL_VER};
 
+RUN rm -rf /etc/apt/sources.list.d/intel-gpu-jammy.list /etc/apt/sources.list.d/oneAPI.list
+
+ENV LD_LIBRARY_PATH=/opt/intel/oneapi/redist/lib:$LD_LIBRARY_PATH
+
+FROM ipex-xpu-base AS ipex-xpu-base-wheels
+
 WORKDIR /
 COPY xpu-requirements.txt .
 
 RUN python -m pip install --no-cache-dir -r xpu-requirements.txt && \
-    rm -rf xpu-requirements.txt /etc/apt/sources.list.d/intel-gpu-jammy.list /etc/apt/sources.list.d/oneAPI.list
-
-ENV LD_LIBRARY_PATH=/opt/intel/oneapi/redist/lib:$LD_LIBRARY_PATH
+    rm -rf xpu-requirements.txt
 
 FROM ipex-xpu-base AS ipex-xpu-jupyter
 
@@ -205,7 +210,8 @@ EXPOSE 8888
 
 CMD ["bash", "-c", "source /etc/bash.bashrc && jupyter notebook --notebook-dir=/jupyter --port 8888 --ip 0.0.0.0 --no-browser --allow-root --ServerApp.token= --ServerApp.password= --ServerApp.allow_origin=* --ServerApp.base_url=$NB_PREFIX"]
 
-FROM ${PYTHON_BASE} as torchserve-base
+
+FROM ${TORCHSERVE_BASE} as torchserve-base
 
 ENV PYTHONUNBUFFERED=TRUE
 
@@ -221,8 +227,6 @@ RUN useradd -m -s /bin/bash model-server && \
     mkdir -p /home/model-server/model-store && \
     chown -R model-server /home/model-server/
 
-FROM torchserve-base AS compile
-
 RUN apt-get update -y && apt-get install -y --no-install-recommends --fix-missing \
     g++ \
     git \
@@ -237,11 +241,6 @@ RUN python3 -m venv /home/venv
 ENV PATH="/home/venv/bin:$PATH"
 
 WORKDIR /home/model-server
-COPY torchserve-requirements.txt .
-COPY requirements.txt .
-
-RUN python -m pip install --no-cache-dir -r requirements.txt && \
-    python -m pip install --no-cache-dir -r torchserve-requirements.txt
 
 RUN echo -e "#!/bin/bash \n\
 set -e \n\
@@ -253,13 +252,29 @@ else \n\
 fi \n\
 tail -f /dev/null" >> /usr/local/bin/dockerd-entrypoint.sh
 
-FROM torchserve-base AS torchserve
+FROM torchserve-base AS compile-cpu
+
+COPY serving/torchserve-requirements.txt .
+COPY requirements.txt .
+
+RUN python -m pip install --no-cache-dir -r requirements.txt && \
+    python -m pip install --no-cache-dir -r torchserve-requirements.txt && \
+    rm -rf requirements.txt torchserve-requirements.txt
+
+FROM torchserve-base AS compile-xpu
+
+COPY serving/torchserve-xpu-requirements.txt .
+
+RUN python -m pip install --no-cache-dir -r torchserve-xpu-requirements.txt && \
+    rm -rf torchserve-xpu-requirements.txt
+
+FROM torchserve-base AS torchserve-cpu
 
 USER model-server
 WORKDIR /home/model-server
 
-COPY --chown=model-server --from=compile /home/venv /home/venv
-COPY --chown=model-server --chmod=755 --from=compile /usr/local/bin/dockerd-entrypoint.sh /usr/local/bin/dockerd-entrypoint.sh
+COPY --chown=model-server --from=compile-cpu /home/venv /home/venv
+COPY --chown=model-server --chmod=755 --from=compile-cpu /usr/local/bin/dockerd-entrypoint.sh /usr/local/bin/dockerd-entrypoint.sh
 COPY --chown=model-server serving/config.properties /home/model-server/config.properties
 
 ENV PATH="/home/venv/bin:$PATH"
@@ -270,3 +285,64 @@ EXPOSE 8080 8081 8082 7070 7071
 
 ENTRYPOINT ["/usr/local/bin/dockerd-entrypoint.sh"]
 CMD ["serve"]
+
+FROM torchserve-base AS torchserve-xpu
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends --fix-missing \
+        gnupg2 \
+        gpg-agent \
+        rsync && \
+    apt-get clean && \
+    rm -rf  /var/lib/apt/lists/*
+
+RUN wget -qO - https://repositories.intel.com/gpu/intel-graphics.key | \
+    gpg --dearmor --yes --output /usr/share/keyrings/intel-graphics.gpg
+RUN echo "deb [arch=amd64 signed-by=/usr/share/keyrings/intel-graphics.gpg] https://repositories.intel.com/gpu/ubuntu jammy unified" | \
+    tee /etc/apt/sources.list.d/intel-gpu-jammy.list
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        jq \
+        curl \
+        libnl-genl-3-200 \
+        intel-gsc \
+        libdrm2 \
+        intel-metrics-discovery \
+        intel-metrics-library && \
+    apt-get autoremove -y && \
+    rm -rf /var/lib/apt/lists/*
+
+ARG XPU_SMI_VERSION
+
+ARG API_URL=https://api.github.com/repos/intel/xpumanager/releases/tags/V${XPU_SMI_VERSION}
+
+RUN wget -q --header="Accept: application/vnd.github.v3+json" --header="User-Agent: MyClient/1.0.0" -O - "$API_URL" | tee /tmp/asset_data.txt && \
+    wget -q --no-check-certificate "$(jq -r '.assets[] | select(.name | test("^xpu-smi.*u22\\.04_amd64\\.deb$")) | .browser_download_url' < /tmp/asset_data.txt)" && \
+    ldconfig && dpkg -i --force-all -- *.deb && \
+    rm -rf -- *.deb /etc/apt/sources.list.d/intel-gpu-jammy.list /etc/apt/sources.list.d/oneAPI.list /tmp/asset_data.txt
+
+ARG GID=109
+
+RUN groupadd -g ${GID} render &&\
+    usermod -aG video,render model-server
+
+USER model-server
+
+WORKDIR /home/model-server
+
+RUN wget --progress=dot:giga https://raw.githubusercontent.com/pytorch/serve/master/examples/intel_extension_for_pytorch/intel_gpu_metric_collector.py && \
+    wget --progress=dot:giga https://raw.githubusercontent.com/pytorch/serve/master/examples/intel_extension_for_pytorch/intel_gpu.py
+
+COPY --chown=model-server --from=compile-xpu /home/venv /home/venv
+COPY --chown=model-server --chmod=755 --from=compile-xpu /usr/local/bin/dockerd-entrypoint.sh /usr/local/bin/dockerd-entrypoint.sh
+COPY --chown=model-server serving/config-xpu.properties /home/model-server/config.properties
+
+ENV PATH="/home/venv/bin:$PATH"
+ENV TEMP=/home/model-server/tmp
+
+# 8080/8081/8082 REST and 7070/7071 gRPC
+EXPOSE 8080 8081 8082 7070 7071
+
+ENTRYPOINT ["/usr/local/bin/dockerd-entrypoint.sh"]
+CMD ["serve"]

pytorch/README.md

@@ -241,8 +241,6 @@ Additionally, if you have a [DeepSpeed* configuration](https://www.deepspeed.ai/
 
 ---
 
-#### Hugging Face Generative AI Container
-
 The image below is an extension of the IPEX Multi-Node Container designed to run Hugging Face Generative AI scripts. The container has the typical installations needed to run and fine tune PyTorch generative text models from Hugging Face. It can be used to run multinode jobs using the same instructions from the [IPEX Multi-Node container](#setup-and-run-ipex-multi-node-container).
 
 | Tag(s)                                | Pytorch  | IPEX         | oneCCL               | HF Transformers | Dockerfile      |
@@ -324,7 +322,7 @@ The images below additionally include [Jupyter Notebook](https://jupyter.org/) s
 
 ## Build from Source
 
-To build the images from source, clone the [Intel® AI Containers](https://github.com/intel/ai-containers) repository, follow the main `README.md` file to setup your environment, and run the following command:
+To build the images from source, clone the [AI Containers](https://github.com/intel/ai-containers) repository, follow the main `README.md` file to setup your environment, and run the following command:
 
 ```bash
 cd pytorch

pytorch/docker-compose.yaml

@@ -122,7 +122,7 @@ services:
         org.opencontainers.base.name: "intel/python:3.10-core"
         org.opencontainers.image.title: "Intel® Extension for PyTorch XPU Base Image"
         org.opencontainers.image.version: ${IPEX_VERSION:-2.1.40}-xpu-${PACKAGE_OPTION:-pip}-base
-      target: ipex-xpu-base
+      target: ipex-xpu-base-wheels
     command: >
       python -c "import torch;print(torch.device('xpu'));import
       intel_extension_for_pytorch as
@@ -156,7 +156,7 @@ services:
     image: ${REGISTRY}/${REPO}:b-${GITHUB_RUN_NUMBER:-0}-${BASE_IMAGE_NAME:-ubuntu}-${BASE_IMAGE_TAG:-22.04}-${PACKAGE_OPTION:-pip}-py${PYTHON_VERSION:-3.10}-ipex-${IPEX_VERSION:-2.1.40xpu}-xpu-jupyter
     ports:
       - 8888:8888
-  torchserve:
+  torchserve-cpu:
     build:
       args:
         PACKAGE_OPTION: pip
@@ -165,22 +165,43 @@ services:
         dependency.apt.openjdk-17-jdk: true
         dependency.idp: false
         dependency.python.ipex: requirements.txt
-        dependency.python.pip: torchserve-requirements.txt
+        dependency.python.pip: serving/torchserve-requirements.txt
         docs: serving
         org.opencontainers.base.name: "intel/python:3.10-core"
         org.opencontainers.image.title: "Intel® Extension for PyTorch Serving Image"
         org.opencontainers.image.version: ${IPEX_VERSION:-2.4.0}-serving-cpu
-      target: torchserve
+      target: torchserve-cpu
     command: torchserve --version
     entrypoint: ""
     extends: ipex-base
-    image: ${REGISTRY}/${REPO}:b-${GITHUB_RUN_NUMBER:-0}-${BASE_IMAGE_NAME:-ubuntu}-${BASE_IMAGE_TAG:-22.04}-py${PYTHON_VERSION:-3.10}-torchserve
+    image: ${REGISTRY}/${REPO}:b-${GITHUB_RUN_NUMBER:-0}-${BASE_IMAGE_NAME:-ubuntu}-${BASE_IMAGE_TAG:-22.04}-py${PYTHON_VERSION:-3.10}-torchserve-cpu
     ports:
       - 8080:8080
       - 8081:8081
       - 8082:8082
       - 7070:7070
       - 7071:7071
+  torchserve-xpu:
+    build:
+      args:
+        PACKAGE_OPTION: pip
+        XPU_SMI_VERSION: ${XPU_SMI_VERSION:-1.2.38}
+        TORCHSERVE_BASE: ipex-xpu-base
+      labels:
+        dependency.apt.numactl: true
+        dependency.apt.openjdk-17-jdk: true
+        dependency.apt.xpu-smi: ${XPU_SMI_VERSION:-1.2.38}
+        dependency.idp: false
+        dependency.python.pip: serving/torchserve-xpu-requirements.txt
+        docs: serving
+        org.opencontainers.base.name: "intel/python:3.10-core"
+        org.opencontainers.image.title: "Intel® Extension for PyTorch XPU Serving Image"
+        org.opencontainers.image.version: ${IPEX_VERSION:-2.1.40}-serving-xpu
+      target: torchserve-xpu
+    command: torchserve --version
+    entrypoint: ""
+    extends: xpu
+    image: ${REGISTRY}/${REPO}:b-${GITHUB_RUN_NUMBER:-0}-${BASE_IMAGE_NAME:-ubuntu}-${BASE_IMAGE_TAG:-22.04}-py${PYTHON_VERSION:-3.10}-torchserve-xpu
   hf-genai:
     build:
       args: