diff --git a/.github/workflows/unittest.yml b/.github/workflows/unittest.yml index 3bac1a99..f9049627 100644 --- a/.github/workflows/unittest.yml +++ b/.github/workflows/unittest.yml @@ -24,41 +24,13 @@ jobs: runs-on: ubuntu-latest strategy: matrix: - config: [pyspark-2.4, tf-1.15, pyarrow-3.0, pyarrow-4.0, latest] + config: [latest] include: - - config: pyspark-2.4 - PYARROW_VERSION: "2.0.0" - NUMPY_VERSION: "1.19.1" - TF_VERSION: "1.15.5" - PYSPARK_VERSION: "2.4.4" - ARROW_PRE_0_15_IPC_FORMAT: 1 - PY: "3.7" - - config: tf-1.15 - PYARROW_VERSION: "2.0.0" - NUMPY_VERSION: "1.19.1" - TF_VERSION: "1.15.5" - PYSPARK_VERSION: "3.0.0" - ARROW_PRE_0_15_IPC_FORMAT: 0 - PY: "3.7" - - config: pyarrow-3.0 - PYARROW_VERSION: "3.0.0" - NUMPY_VERSION: "1.19.1" - TF_VERSION: "2.5.0" - PYSPARK_VERSION: "3.0.0" - ARROW_PRE_0_15_IPC_FORMAT: 0 - PY: "3.7" - - config: pyarrow-4.0 - PYARROW_VERSION: "4.0.0" - NUMPY_VERSION: "1.19.1" - TF_VERSION: "2.5.0" - PYSPARK_VERSION: "3.0.0" - ARROW_PRE_0_15_IPC_FORMAT: 0 - PY: "3.7" - config: latest PYARROW_VERSION: "6.0.1" NUMPY_VERSION: "1.21.5" TF_VERSION: "2.8.0" - PYSPARK_VERSION: "3.0.0" + PYSPARK_VERSION: "3.4.0" ARROW_PRE_0_15_IPC_FORMAT: "0" PY: "3.9" @@ -91,7 +63,8 @@ jobs: $RUN pip install -U pip setuptools $RUN pip install -e /petastorm/[test,tf,torch,docs,opencv] $RUN pip install --upgrade numpy==$NUMPY_VERSION - $RUN pip install -U pyarrow==${PYARROW_VERSION} tensorflow==${TF_VERSION} pyspark==${PYSPARK_VERSION} + $RUN pip install -U pyarrow==${PYARROW_VERSION} tensorflow==${TF_VERSION} + $RUN pip install https://dist.apache.org/repos/dist/dev/spark/v3.4.0-rc1-bin/pyspark-3.4.0.tar.gz $RUN pip list $RUN mypy petastorm $RUN flake8 . --count --show-source --statistics diff --git a/docker/Dockerfile b/docker/Dockerfile index f358cd69..5045c547 100644 --- a/docker/Dockerfile +++ b/docker/Dockerfile @@ -43,7 +43,8 @@ RUN /petastorm_venv3.7/bin/pip3.7 install --no-cache scikit-build RUN /petastorm_venv3.7/bin/pip3.7 install --no-cache -e /petastorm/[test,tf,torch,docs,opencv] --only-binary pyarrow --only-binary opencv-python RUN /petastorm_venv3.7/bin/pip3.7 uninstall -y petastorm # To avoid some version incompatibilities, we pin these libraries to versions that known to work together -RUN /petastorm_venv3.7/bin/pip3.7 install -U pyarrow==3.0.0 numpy==1.19.1 tensorflow==2.1.0 pyspark==3.0.0 +RUN /petastorm_venv3.7/bin/pip3.7 install -U pyarrow==3.0.0 numpy==1.19.1 tensorflow==2.1.0 +RUN /petastorm_venv3.7/bin/pip3.7 install https://dist.apache.org/repos/dist/dev/spark/v3.4.0-rc1-bin/pyspark-3.4.0.tar.gz # Otherwise we might have trouble with loading of libGL.so.1 RUN /petastorm_venv3.7/bin/pip3.7 install opencv-python-headless @@ -58,7 +59,8 @@ RUN /petastorm_venv3.9/bin/pip3.9 install --no-cache scikit-build RUN /petastorm_venv3.9/bin/pip3.9 install --no-cache -e /petastorm/[test,tf,torch,docs,opencv] --only-binary pyarrow --only-binary opencv-python RUN /petastorm_venv3.9/bin/pip3.9 uninstall -y petastorm # To avoid some version incompatibilities, we pin these libraries to versions that known to work together -RUN /petastorm_venv3.9/bin/pip3.9 install -U pyarrow==3.0.0 numpy==1.19.3 tensorflow==2.5.0 pyspark==3.0.0 +RUN /petastorm_venv3.9/bin/pip3.9 install -U pyarrow==3.0.0 numpy==1.19.3 tensorflow==2.5.0 +RUN /petastorm_venv3.7/bin/pip3.7 install https://dist.apache.org/repos/dist/dev/spark/v3.4.0-rc1-bin/pyspark-3.4.0.tar.gz # Otherwise we might have trouble with loading of libGL.so.1 RUN /petastorm_venv3.9/bin/pip3.9 install opencv-python-headless