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
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	LABEL	`com.nvidia.build.ref=bb0476c6ce5f374d5bb697427f59dd429c527a37`	08/07/2023 12:21 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	ARG	`NVIDIA_BUILD_REF`	08/07/2023 12:21 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	LABEL	`com.nvidia.build.id=66128610`	08/07/2023 12:21 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	ENV	`NVIDIA_BUILD_ID=66128610`	08/07/2023 12:21 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	ARG	`NVIDIA_BUILD_ID`	08/07/2023 12:21 AM UTC
sha256:756610a546ddbb6b149e49aeb67381d64929f093baa84d4e5383e894fb60732a	COPY	`entrypoint.d/ /opt/nvidia/entrypoint.d/`	08/07/2023 12:21 AM UTC
sha256:7e63c04b708a3fbd958517608015e30f7098452f74a0b5bd4ebad0e5fe1fcbc1	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 ${_CUDA_COMPAT_PATH}/lib`	08/07/2023 12:21 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	ENV	`CUDA_MODULE_LOADING=LAZY`	08/07/2023 12:21 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	ENV	`TORCH_CUDNN_V8_API_ENABLED=1`	08/07/2023 12:21 AM UTC
sha256:86ea29fc5964c78eb93fc15cced07544d0a8372879681c1fe71bd265b11e5df7	RUN	`PYVER=3.10 env MAX_JOBS=1 pip install --no-cache-dir --no-build-isolation git+https://github.com/NVIDIA/TransformerEngine.git@release_v${TRANSFORMER_ENGINE_VERSION}`	08/07/2023 12:21 AM UTC