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
a9b3c7d6f208462540b93852e6f8ca7cc0447a41f1d5ecc7bdd40eb1ea21fb80	CONFIG	`Entrypoint /opt/nvidia/nvidia_entrypoint.sh; WorkingDir /workspace; ExposedPorts 6006/tcp, 8888/tcp`	05/20/2023 4:25 AM UTC
a18306692113d4f08f21f9361988d494fe8a97e4eee3dbe7c0b83a8c43be3141	ARG	`NVIDIA_BUILD_REF`	05/20/2023 4:25 AM UTC
a5557f24a26672417b8c0e2123af8258b12b0b481f96338fc648311840706a67	LABEL	`com.nvidia.build.id=60708168`	05/20/2023 4:25 AM UTC
77b18f69e8620f510a45cccfd566fb74f3b18f7a89d9aab2d2ca060eb61c6b53	ENV	`NVIDIA_BUILD_ID=60708168`	05/20/2023 4:25 AM UTC
b4e3a88605b89b64e7e9c018915e533bdc50276569cf49e486a14f2f86c017bd	ARG	`NVIDIA_BUILD_ID`	05/20/2023 4:25 AM UTC
dedf21d0f2cd92ca2572f35821a4e631a51015745ee6271db9847cc87e1b184c	COPY	`entrypoint.d/ /opt/nvidia/entrypoint.d/`	05/20/2023 4:25 AM UTC
e0c7e16a7a825349fdda08acaf01ef6dc07e704106c8e8f0d280940dcb21e2bd	RUN	`PYVER=3.10 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 ${_CUDA_COMPAT_PATH}/lib`	05/20/2023 4:25 AM UTC
756772dd518fbe2fefd4432b0119e0977f80c0e80a7fad1d227b65a8772755bf	ENV	`CUDA_MODULE_LOADING=LAZY`	05/20/2023 4:25 AM UTC
0cd6eb8738bc8ede2f0e23a9cb435e3c94193744ec37c15c79cf9b300059ee8d	ENV	`TORCH_CUDNN_V8_API_ENABLED=1`	05/20/2023 4:25 AM UTC
c221c50709ea67442e58c414a7698d15eb75e30f8065dddf4ab294bf2c46884a	RUN	`PYVER=3.10 pip install --no-cache-dir git+https://github.com/NVIDIA/TransformerEngine.git@release_v${TRANSFORMER_ENGINE_VERSION}`	05/20/2023 4:25 AM UTC