Merge pull request #162 from aws/numpy

Numpy Installation Issue & ARM Dockerfile PyArrow Update

Author: amzn-choeric

docker/1.0-1/base/Dockerfile.cpu

@@ -10,7 +10,6 @@ ARG CONDA_PKG_VERSION=4.13.0
 ARG PYTHON_VERSION=3.8.13
 ARG PYARROW_VERSION=10.0.1
 ARG MLIO_VERSION=v0.8.0
-ARG NUMPY_VERSION=1.24.1
 
 # Install python and other scikit-learn runtime dependencies
 # Dependency list from http://scikit-learn.org/stable/developers/advanced_installation.html#installing-build-dependencies
@@ -82,7 +81,6 @@ RUN echo "conda ${CONDA_PKG_VERSION}" >> /miniconda3/conda-meta/pinned && \
     conda install -c conda-forge python=${PYTHON_VERSION} && \
     conda install conda=${CONDA_PKG_VERSION} && \
     conda update -y conda && \
-    conda install -c conda-forge numpy=${NUMPY_VERSION} && \
     conda install -c conda-forge pyarrow=${PYARROW_VERSION} && \
     cd /tmp && \
     git clone --branch ${MLIO_VERSION} https://github.com/awslabs/ml-io.git mlio && \

docker/1.0-1/base/Dockerfile_arm.cpu

@@ -7,8 +7,8 @@ ARG MINICONDA_VERSION=4.9.2 # Upgraded version
 ARG CONDA_PY_VERSION=38
 ARG CONDA_PKG_VERSION=4.10.1 
 ARG PYTHON_VERSION=3.8.13
-ARG PYARROW_VERSION=1.0.0
-ARG MLIO_VERSION=arch-agnostic
+ARG PYARROW_VERSION=10.0.1
+ARG MLIO_VERSION=0.8.0
 
 # Install python and other scikit-learn runtime dependencies
 # Dependency list from http://scikit-learn.org/stable/developers/advanced_installation.html#installing-build-dependencies
@@ -66,6 +66,10 @@ RUN apt-get update && \
     rm /etc/apt/trusted.gpg.d/kitware.gpg && \
     rm -rf /var/lib/apt/lists/*
 
+# http://ftp.us.debian.org/debian/pool/main/libf/libffi/libffi7_3.3-6_arm64.deb
+COPY docker/1.0-1/resources/libffi7_3.3-6_arm64.deb /tmp
+RUN dpkg -i /tmp/libffi7_3.3-6_arm64.deb
+
 RUN cd /tmp && \
     curl -L --output /tmp/Miniconda3.sh https://repo.anaconda.com/miniconda/Miniconda3-py${CONDA_PY_VERSION}_${MINICONDA_VERSION}-Linux-aarch64.sh && \
     bash /tmp/Miniconda3.sh -bfp /miniconda3 && \

docker/1.0-1/final/Dockerfile.cpu

@@ -8,9 +8,9 @@ RUN python -m pip install -r /requirements.txt && \
     rm /requirements.txt
 
 COPY dist/sagemaker_sklearn_container-2.0-py3-none-any.whl /sagemaker_sklearn_container-2.0-py3-none-any.whl
-# https://github.com/googleapis/google-cloud-python/issues/6647
-RUN rm -rf /miniconda3/lib/python3.7/site-packages/numpy-1.19.4.dist-info && \
-    pip install --no-cache /sagemaker_sklearn_container-2.0-py3-none-any.whl && \
+RUN rm /miniconda3/lib/python3.8/site-packages/**/REQUESTED && \
+    rm /miniconda3/lib/python3.8/site-packages/**/direct_url.json
+RUN pip install --no-cache /sagemaker_sklearn_container-2.0-py3-none-any.whl && \
     rm /sagemaker_sklearn_container-2.0-py3-none-any.whl
 
 ENV SAGEMAKER_TRAINING_MODULE sagemaker_sklearn_container.training:main

docker/1.0-1/resources/libffi7_3.3-6_arm64.deb

@@ -22,3 +22,4 @@ MarkupSafe==2.1.1
 Werkzeug==0.15.6
 setuptools==65.5.1
 wheel==0.38.1
+numpy==1.24.1