PyTorch | NVIDIA NGC

NVIDIA

PyTorch

Container

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.

NVIDIA AI Enterprise Supported

Layer	Label		Created
3993969b4db5a98b269575513004949c60136d3f2328b068b58dc65e51bdfe8a	CONFIG	`Entrypoint /opt/nvidia/nvidia_entrypoint.sh; WorkingDir /workspace; ExposedPorts 6006/tcp, 8888/tcp`	09/22/2022 10:03 PM UTC
85e197f27d3ba281f306797492239865f0340b95c3054c9cfd5361d6bf6ff2cb	ARG	`NVIDIA_BUILD_REF`	09/22/2022 10:03 PM UTC
ce843cbcdfed9e0d2d25294de03ef7a3054e86f35d85bf65e3062cd3d6997d58	LABEL	`com.nvidia.build.id=44877844`	09/22/2022 10:03 PM UTC
eaa4f8769fe03205c628d34d9014237b0b13428605dc6378148cf75af50c49a2	ENV	`NVIDIA_BUILD_ID=44877844`	09/22/2022 10:03 PM UTC
73b69ba7954939acfdc8ed9512f1e4cc16d7159bccd273d9f7ff662c6ac7fb31	ARG	`NVIDIA_BUILD_ID`	09/22/2022 10:03 PM UTC
8db20a3eb63e1a923396b9cbbe2010c5259308c0fd8c5bf2cccfb2e8bb7bab99	COPY	`entrypoint.d/ /opt/nvidia/entrypoint.d/`	09/22/2022 10:03 PM UTC
2939a04b2fc051bd4e5c094e7fa3cc39605980a32a6b8329db1f93c7610c57ef	RUN	`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`	09/22/2022 10:03 PM UTC
cfbe5e39502eede50f678c6092d772c2cb34aa791715aaab0c34bac442c59b78	ENV	`CUDA_MODULE_LOADING=LAZY`	09/22/2022 10:02 PM UTC
cd4a0658192870da46f095f0617c5637129ea462f51c42988f1255f57e9cf9da	ENV	`TORCH_CUDNN_V8_API_ENABLED=1`	09/22/2022 10:02 PM UTC
ad47b9d8861c7ff34684370a3729725166c9d92e1143dbe7097569cd1545550f	RUN	`PYVER=3.8 pip install --no-cache-dir /tmp/te_wheel/*.whl`	09/22/2022 10:02 PM UTC