NVIDIA
NVIDIA
PyTorch
Container
NVIDIA
NVIDIA
PyTorch

PyTorch is a GPU accelerated tensor computational framework. Functionality can be extended with common Python libraries such as NumPy and SciPy. Automatic differentiation is done with a tape-based system at the functional and neural network layer levels.

LayerLabelCreated
7321c1ba9e4830ae7f9d8c7d2dbf82912bfc74e58895c0477178a3e8a064aa3cCONFIG
Entrypoint /opt/nvidia/nvidia_entrypoint.sh; WorkingDir /workspace; ExposedPorts 6006/tcp, 8888/tcp
10/10/2021 12:37 AM UTC
be5ef0e58cdfe2ccf09bf6bc03b95f89a5691c5aa96b7cb23ea8b0eccdc32fe2ARG
NVIDIA_BUILD_REF
10/10/2021 12:37 AM UTC
cacd9523da8946f75e3cb3b097dcb51494ec9f69a22ed4c22b56f8ab2bff5dc2LABEL
com.nvidia.build.id=28019337
10/10/2021 12:37 AM UTC
3417b208d89965fdd79ab0b6f18f8b20e3955900bbdd5a1e70d24f65b0ae222dENV
NVIDIA_BUILD_ID=28019337
10/10/2021 12:37 AM UTC
9d44f11a1f35858e30ff51066b799d546b27803678b05e8e7d575c34d7802e4eARG
NVIDIA_BUILD_ID
10/10/2021 12:37 AM UTC
6a297600ab02fafbeca51c6eaf2aa15f0a10b3db73457c45ffe8b80191daa07aCOPY
entrypoint.d/ /opt/nvidia/entrypoint.d/
10/10/2021 12:37 AM UTC
342f959721a79fb133437afe4cfa430517e93b944149820ab7e8c6d71e049e35RUN
NVIDIA_PYTORCH_VERSION=21.10 PYTORCH_BUILD_VERSION=1.10.0a0+0aef44c TARGETARCH=amd64 PYVER=3.8 ln -sf ${_CUDA_COMPAT_PATH}/lib.real ${_CUDA_COMPAT_PATH}/lib &&
  echo ${_CUDA_COMPAT_PATH}/lib > /etc/ld.so.conf.d/00-cuda-compat.conf &&
  ldconfig &&
  rm -f ${_CUDA_COMPAT_PATH}/lib
10/10/2021 12:37 AM UTC
acd260219a34a6b1e4c77a410e6d7662f84a95e09bf04adfdfdcd02fdbea8e07RUN
NVIDIA_PYTORCH_VERSION=21.10 PYTORCH_BUILD_VERSION=1.10.0a0+0aef44c TARGETARCH=amd64 PYVER=3.8 mkdir -p /opt/dlprof &&
  mkdir -p /opt/dlprof_viewer_install &&
  cp /nvidia/tmp/pip/* /opt/dlprof_viewer_install/ &&
  cp /nvidia/opt/dlprof/bin/dlprof /opt/dlprof/ &&
  cp /nvidia/workspace/LICENSE /opt/dlprof/ &&
  cp /nvidia/workspace/README.md /opt/dlprof/ &&
  ln -sf /opt/dlprof/dlprof /usr/local/bin/dlprof &&
  cd /nvidia/opt/dlprof/bin/ &&
  pip install --no-cache-dir nvidia_dlprof_pytorch_nvtx* &&
  cd /opt/dlprof_viewer_install &&
  pip install --no-cache-dir nvidia_dlprofviewer-* &&
  rm -rf /opt/conda/lib/python3.8/site-packages/qa
10/10/2021 12:36 AM UTC
b6f3d8d95fe29e4ed0f10e3974c103dfa2964cfb4d4b7ed7415615ddc5798731RUN
NVIDIA_PYTORCH_VERSION=21.10 PYTORCH_BUILD_VERSION=1.10.0a0+0aef44c TARGETARCH=amd64 PYVER=3.8 pip --version &&
  python -c 'import sys; print(sys.platform)' &&
  pip install --no-cache-dir nvidia-pyindex &&
  pip install --extra-index-url https://urm.nvidia.com/artifactory/api/pypi/sw-tensorrt-pypi/simple --no-cache-dir polygraphy==0.33.0 &&
  if [[ $TARGETARCH = "amd64" ]] ; then pip install --extra-index-url http://sqrl/dldata/pip-simple --trusted-host sqrl --no-cache-dir pytorch-quantization==2.1.0; fi
10/10/2021 12:35 AM UTC
c0d19a2922dc973b7a2bb161fc5bce2e85ab49997b40cee14854238cd3aea0a0ENV
PATH=/opt/conda/bin:/opt/cmake-3.19.3-Linux-$(uname -m)/bin/:/usr/local/mpi/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/ucx/bin:/opt/tensorrt/bin
10/10/2021 12:35 AM UTC
...