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
4dc6bd6036afc512a3720c22ec7cb4a4be86ee5ec5c97f0a08cdeebf4a9b7f7b	CONFIG	`Entrypoint /opt/nvidia/nvidia_entrypoint.sh; WorkingDir /workspace; ExposedPorts 6006/tcp, 8888/tcp`	01/28/2023 12:35 PM UTC
268713e28beb9cc570c6332f522c45a100575d945bcc6361fa9ac117d99a0d6e	ARG	`NVIDIA_BUILD_REF`	01/28/2023 12:35 PM UTC
636dbb311e78568deaf564327f48731cc76c2747a92d0c3e644c262b6a81443d	LABEL	`com.nvidia.build.id=52269074`	01/28/2023 12:35 PM UTC
ae94bc456396add8e257cb9d9d30f9059fa7c950de6fdabe24cf7067831e4251	ENV	`NVIDIA_BUILD_ID=52269074`	01/28/2023 12:35 PM UTC
d51b99f602d33e9bb8f81c7768061382ced853e57a7ab8985606cb0f01f99421	ARG	`NVIDIA_BUILD_ID`	01/28/2023 12:35 PM UTC
b85d9519c0ae831d1a0838871313c5d4083784556ce69f2e1a28e693ceb21f5b	COPY	`entrypoint.d/ /opt/nvidia/entrypoint.d/`	01/28/2023 12:35 PM UTC
0a845db197107f02dc245b45002c22efdd763585a93f765e600bbd8e245873bd	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 ${_CUDA_COMPAT_PATH}/lib`	01/28/2023 12:35 PM UTC
0579131894539208eb5f861adae453a11666bb99e10fd966c39153f5c1f422a0	ENV	`CUDA_MODULE_LOADING=LAZY`	01/28/2023 12:35 PM UTC
6a88601fe97f0e66c971fed7cfbd8b22b514ae4157fea38a3c0c67ef9785c8b7	ENV	`TORCH_CUDNN_V8_API_ENABLED=1`	01/28/2023 12:35 PM UTC
487c2ecb65112a42f8b2d9b42663bdf3df8a2b066086386afa2ea597b92f83ea	RUN	`PYVER=3.8 pip install --no-cache-dir git+https://github.com/NVIDIA/TransformerEngine.git@release_v${TRANSFORMER_ENGINE_VERSION}`	01/28/2023 12:35 PM UTC