diff --git a/README.md b/README.md index 9e4ccc482c..50f6d32a09 100644 --- a/README.md +++ b/README.md @@ -82,7 +82,7 @@ Get started with our [3 steps guide](https://pytorch-lightning.readthedocs.io/en | System / PyTorch ver. | 1.3 (min. req.)* | 1.4 | 1.5 | 1.6 (latest) | 1.7 (nightly) | | :---: | :---: | :---: | :---: | :---: | :---: | | Conda py3.7 [linux] | [![PyTorch & Conda](https://github.com/PyTorchLightning/pytorch-lightning/workflows/PyTorch%20&%20Conda/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22PyTorch+%26+Conda%22+branch%3Amaster) | [![PyTorch & Conda](https://github.com/PyTorchLightning/pytorch-lightning/workflows/PyTorch%20&%20Conda/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22PyTorch+%26+Conda%22+branch%3Amaster) | [![PyTorch & Conda](https://github.com/PyTorchLightning/pytorch-lightning/workflows/PyTorch%20&%20Conda/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22PyTorch+%26+Conda%22+branch%3Amaster) | [![PyTorch & Conda](https://github.com/PyTorchLightning/pytorch-lightning/workflows/PyTorch%20&%20Conda/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22PyTorch+%26+Conda%22+branch%3Amaster) | [![PyTorch & Conda](https://github.com/PyTorchLightning/pytorch-lightning/workflows/PyTorch%20&%20Conda/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22PyTorch+%26+Conda%22+branch%3Amaster) | -| Linux py3.7 [GPUs**] | - | - | - | [![Build Status](http://35.192.60.23/api/badges/PyTorchLightning/pytorch-lightning/status.svg)](http://35.192.60.23/PyTorchLightning/pytorch-lightning) | - | +| Linux py3.7 [GPUs**] | - | - |[![Build Status](http://104.154.220.231/api/badges/PyTorchLightning/pytorch-lightning/status.svg)](http://104.154.220.231/PyTorchLightning/pytorch-lightning) | - | - | | Linux py3.7 [TPUs***] | - | - | - | [![TPU tests](https://github.com/PyTorchLightning/pytorch-lightning/workflows/TPU%20tests/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22TPU+tests%22+branch%3Amaster) | - | | Linux py3.6 / py3.7 / py3.8 | [![CI complete testing](https://github.com/PyTorchLightning/pytorch-lightning/workflows/CI%20complete%20testing/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22CI+testing%22) | - | - | [![CI complete testing](https://github.com/PyTorchLightning/pytorch-lightning/workflows/CI%20complete%20testing/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22CI+testing%22) | - | | OSX py3.6 / py3.7 | - | [![CI complete testing](https://github.com/PyTorchLightning/pytorch-lightning/workflows/CI%20complete%20testing/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22CI+testing%22) | - | [![CI complete testing](https://github.com/PyTorchLightning/pytorch-lightning/workflows/CI%20complete%20testing/badge.svg)](https://github.com/PyTorchLightning/pytorch-lightning/actions?query=workflow%3A%22CI+testing%22) | - | diff --git a/dockers/base-cuda/Dockerfile b/dockers/base-cuda/Dockerfile index 75dec1906a..2408fb5570 100644 --- a/dockers/base-cuda/Dockerfile +++ b/dockers/base-cuda/Dockerfile @@ -27,29 +27,22 @@ RUN apt-get update && apt-get install -y --no-install-recommends \ curl \ ca-certificates \ && \ - # Cleaning + +# Install conda and python. +# NOTE new Conda does not forward the exit status... https://github.com/conda/conda/issues/8385 + curl -o ~/miniconda.sh https://repo.anaconda.com/miniconda/Miniconda3-4.7.12-Linux-x86_64.sh && \ + chmod +x ~/miniconda.sh && \ + ~/miniconda.sh -b && \ + rm ~/miniconda.sh && \ + +# Cleaning apt-get autoremove -y && \ apt-get clean && \ rm -rf /root/.cache && \ rm -rf /var/lib/apt/lists/* -# add non-root user -RUN useradd --create-home --shell /bin/bash flash - -USER flash -ENV CONDA_ENV=lightning -ENV WORKDIR=/home/flash -WORKDIR $WORKDIR - -# install conda and python -RUN curl -o ~/miniconda.sh https://repo.anaconda.com/miniconda/Miniconda3-${CONDA_VERSION}-Linux-x86_64.sh && \ - chmod +x ~/miniconda.sh && \ - ~/miniconda.sh -b -p ${WORKDIR}/miniconda && \ - rm ~/miniconda.sh - -# add conda to path -ENV PATH="${WORKDIR}/miniconda/bin:$PATH" -ENV LD_LIBRARY_PATH="${WORKDIR}/miniconda/lib:$LD_LIBRARY_PATH" +ENV PATH="/root/miniconda3/bin:$PATH" +ENV LD_LIBRARY_PATH="/root/miniconda3/lib:$LD_LIBRARY_PATH" ENV CUDA_TOOLKIT_ROOT_DIR="/usr/local/cuda" ENV HOROVOD_GPU_OPERATIONS=NCCL @@ -59,46 +52,48 @@ ENV HOROVOD_WITHOUT_MXNET=1 ENV HOROVOD_WITH_GLOO=1 ENV HOROVOD_WITHOUT_MPI=1 -COPY --chown=flash environment.yml environment.yml +ENV CONDA_ENV=lightning +COPY environment.yml environment.yml # conda init -RUN conda create -y --name $CONDA_ENV python=$PYTHON_VERSION pytorch=$PYTORCH_VERSION torchvision cudatoolkit=$CUDA_VERSION --channel=$PYTORCH_CHANNEL && \ +RUN conda create -y --name $CONDA_ENV "python=$PYTHON_VERSION" "pytorch=$PYTORCH_VERSION" torchvision "cudatoolkit=$CUDA_VERSION" --channel=$PYTORCH_CHANNEL && \ conda init bash && \ # NOTE: this requires that the channel is presented in the yaml before packages - python -c "fname = 'environment.yml' ; req = open(fname).read().replace('pytorch', '${PYTORCH_CHANNEL}', 1) ; open(fname, 'w').write(req)" && \ - python -c "fname = 'environment.yml' ; req = open(fname).readlines() ; open(fname, 'w').writelines([l for l in req if 'horovod' not in l])" && \ + # replace channel to nigtly if neede, fix PT version and remove Horovod as it will be installe later + python -c "import re ; fname = 'environment.yml' ; req = re.sub(r'torch>=[\d\.]+', 'torch=${PYTORCH_VERSION}', open(fname).read().replace('pytorch', '${PYTORCH_CHANNEL}', 1)) ; open(fname, 'w').writelines([l for l in req if 'horovod' not in l])" && \ + cat environment.yml && \ conda env update --file environment.yml && \ conda clean -ya && \ # conda config --set auto_activate_base true && \ rm environment.yml -ENV PATH=${WORKDIR}/miniconda/envs/${CONDA_ENV}/bin:$PATH -ENV LD_LIBRARY_PATH="${WORKDIR}/miniconda/envs/${CONDA_ENV}/lib:$LD_LIBRARY_PATH" +ENV PATH /root/miniconda3/envs/${CONDA_ENV}/bin:$PATH +ENV LD_LIBRARY_PATH="/root/miniconda3/envs/${CONDA_ENV}/lib:$LD_LIBRARY_PATH" # if you want this environment to be the default one, uncomment the following line: -#ENV CONDA_DEFAULT_ENV=${CONDA_ENV} -#ENV CONDA_AUTO_ACTIVATE_BASE=true +ENV CONDA_DEFAULT_ENV=${CONDA_ENV} -# set default init conda env, the cases above with ENV did not work... -RUN echo ". ${WORKDIR}/miniconda/etc/profile.d/conda.sh" >> ~/.bashrc && \ - echo "conda activate ${CONDA_ENV}" >> ~/.bashrc - -COPY --chown=flash ./requirements/extra.txt requirements-extra.txt -COPY --chown=flash ./requirements/test.txt requirements-tests.txt -COPY --chown=flash ./requirements/examples.txt requirements-examples.txt +COPY ./requirements/extra.txt requirements-extra.txt +COPY ./requirements/test.txt requirements-tests.txt +COPY ./requirements/examples.txt requirements-examples.txt RUN \ + # Disable cache + pip config set global.cache-dir false && \ + #echo ". ${WORKDIR}/miniconda/etc/profile.d/conda.sh" >> ~/.bashrc && \ + #echo "conda activate ${CONDA_ENV}" >> ~/.bashrc && \ + #source ~/.bashrc && \ # filter only Horovod python -c "fname = 'requirements-extra.txt' ; req = open(fname).readlines() ; open(fname, 'w').writelines([l for l in req if 'horovod' in l])" && \ - #cat requirements-extra.txt && \ # Install all requirements MAKEFLAGS="-j$(nproc)" ; pip install -r requirements-extra.txt && \ - pip install -r requirements-tests.txt --upgrade-strategy only-if-needed --no-cache-dir && \ - pip install -r requirements-examples.txt --upgrade-strategy only-if-needed --no-cache-dir && \ - rm requirements* && \ + pip install -r requirements-tests.txt --upgrade-strategy only-if-needed && \ + pip install -r requirements-examples.txt --upgrade-strategy only-if-needed && \ + rm requirements* + +RUN \ # Show what we have pip --version && \ conda info && \ conda list && \ - pip list - -CMD ["/bin/bash"] + pip list && \ + python -c "import torch; print(torch.__version__)" \ No newline at end of file diff --git a/dockers/base-xla/Dockerfile b/dockers/base-xla/Dockerfile index 2af21206d3..68d5666fcd 100644 --- a/dockers/base-xla/Dockerfile +++ b/dockers/base-xla/Dockerfile @@ -35,14 +35,15 @@ RUN apt-get update && \ # Cleaning apt-get autoremove -y && \ apt-get clean && \ - rm -rf /root/.cache + rm -rf /root/.cache && \ + rm -rf /var/lib/apt/lists/* ENV PATH="/root/miniconda3/bin:$PATH" ENV LD_LIBRARY_PATH="/root/miniconda3/lib:$LD_LIBRARY_PATH" -RUN conda create -y --name $CONDA_ENV python=$PYTHON_VERSION && \ +RUN conda create -y --name $CONDA_ENV "python=$PYTHON_VERSION" && \ conda init bash && \ - conda install -y python=$PYTHON_VERSION mkl && \ + conda install -y mkl && \ # Disable cache pip config set global.cache-dir false && \