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
35d11422e08663b1dea0a5cebcd1a46b15611bf8c2bbed0aa1bc0b5f0ceca28eCONFIG
Entrypoint /usr/local/bin/nvidia_entrypoint.sh; WorkingDir /workspace; ExposedPorts 6006/tcp, 8888/tcp
07/23/2021 8:40 AM UTC
aead81730f461c1a6042cbc7b4a23419d1c33d588d3837b14fc4235f94bf6d9fARG
NVIDIA_BUILD_REF
07/23/2021 8:40 AM UTC
b4f58c221d2f05b87d59913f2632d55b0f02f6c52ca4832836ceac81bc1ba814LABEL
com.nvidia.build.id=25165078
07/23/2021 8:40 AM UTC
33918089d505dcff7f7322ce902f8eb9893af426d2dcadf27258fab921205a1eENV
NVIDIA_BUILD_ID=25165078
07/23/2021 8:40 AM UTC
f3eec0e62f6273d408d0e5f86509b6a4e56dac5cb4baa0a95b5a505f1f649867ARG
NVIDIA_BUILD_ID
07/23/2021 8:40 AM UTC
02b1de15f00e0efc7de71f9964c5d923c920a012f8d5a757ce98bec7a1b059b4RUN
NVIDIA_PYTORCH_VERSION=21.07 PYTORCH_BUILD_VERSION=1.10.0a0+ecc3718 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
07/23/2021 8:40 AM UTC
5eab4e1b2c3a4c43a709c6ff7e9585e74dc7a9cc20af64badf3d8cdd0ca92b6eRUN
NVIDIA_PYTORCH_VERSION=21.07 PYTORCH_BUILD_VERSION=1.10.0a0+ecc3718 PYVER=3.8 pip install --no-cache-dir --extra-index-url https://pypi.ngc.nvidia.com --extra-index-url https://tensorrt-read-only:Tensorrt\@123@urm.nvidia.com/artifactory/api/pypi/sw-tensorrt-pypi/simple polygraphy pytorch-quantization
07/23/2021 8:40 AM UTC
f1d1c13a4376884bf30580272216f7d543dfe0d9c9341cea7a34ff30a1b562b5ENV
PATH=/opt/conda/bin:/opt/cmake-3.14.6-Linux-x86_64/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
07/23/2021 8:40 AM UTC
ed607867f2aee97a97c4f4f3b9b3db1d35e52499635aa4432fa353d1b7a00b9aRUN
NVIDIA_PYTORCH_VERSION=21.07 PYTORCH_BUILD_VERSION=1.10.0a0+ecc3718 PYVER=3.8 URL=$(VERIFY=1 /nvidia/build-scripts/installTRT.sh 2>/dev/null | sed -n "s/^.*\(http.*\)deb.*$/\1/p")tar &&
  FILE=$(wget -O - $URL 2>/dev/null | sed -n 's/^.*href="\(TensorRT[^"]*\)".*$/\1/p' | grep -v internal) &&
  wget --quiet $URL/$FILE -O - | tar -xz &&
  PY=$(python -c 'import sys; print(str(sys.version_info[0])+str(sys.version_info[1]))') &&
  pip install TensorRT-*/python/tensorrt-*-cp$PY*.whl &&
  pip install TensorRT-*/graphsurgeon/graphsurgeon-*.whl &&
  pip install TensorRT-*/uff/uff-*.whl &&
  mv /usr/src/tensorrt /opt &&
  ln -s /opt/tensorrt /usr/src/tensorrt &&
  rm -r TensorRT-* &&
  UFF_PATH=$(pip show uff | sed -n 's/Location: \(.*\)$/\1/p')/uff &&
  sed -i 's/from tensorflow import GraphDef/from tensorflow.python import GraphDef/' $UFF_PATH/converters/tensorflow/conversion_helpers.py &&
  chmod +x ${UFF_PATH}/bin/convert_to_uff.py &&
  ln -sf ${UFF_PATH}/bin/convert_to_uff.py /usr/local/bin/convert-to-uff
07/23/2021 8:40 AM UTC
a0e8792c620a7fae03ce449073058abbdd1d5da4c8e13d1c90dd082522c97ff2COPY
NVIDIA_Deep_Learning_Container_License.pdf /workspace/
07/23/2021 8:40 AM UTC
...