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
d447214f89c6db082b737985660179923668edc1ea03f5a952abd66d508dfee2	CONFIG	`Entrypoint /opt/nvidia/nvidia_entrypoint.sh; WorkingDir /workspace; ExposedPorts 6006/tcp, 8888/tcp`	03/16/2023 1:35 AM UTC
14072722997e3117b9c26f875d0a26ef429abfcadfd26eec08906d8de90aeab7	ARG	`NVIDIA_BUILD_REF`	03/16/2023 1:35 AM UTC
5c7caa4b1893be7d321e4271db6f877d58bb1d691127a0ed3faad32b614d0c94	LABEL	`com.nvidia.build.id=55416458`	03/16/2023 1:35 AM UTC
d66fe4289f9d1b084071db032b6d9233acabd5f39b4496df481db201bf7cfc90	ENV	`NVIDIA_BUILD_ID=55416458`	03/16/2023 1:35 AM UTC
473403385dbd8af6ffcb182eae488e9fff0b389170235844cf74be17f00d6a23	ARG	`NVIDIA_BUILD_ID`	03/16/2023 1:35 AM UTC
4236172898ac29cdd5ec970be11a3d97bd8d7392abdde229e302d3c13548fa29	COPY	`entrypoint.d/ /opt/nvidia/entrypoint.d/`	03/16/2023 1:35 AM UTC
98fffbf357b862ae777100da122991972ba9ab0c3ad3d168277a214b7fb94a29	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 ${_CUDA_COMPAT_PATH}/lib`	03/16/2023 1:35 AM UTC
7eb1f7cb116ebf29be33fefa9d64e1a9148f1c92c6f77eb5c0c3a53b0699cb4b	ENV	`CUDA_MODULE_LOADING=LAZY`	03/16/2023 1:35 AM UTC
208ac7e54b890edbc6e00222e15e5ec20e4a58bcf0494f0f51a1d501346c07f1	ENV	`TORCH_CUDNN_V8_API_ENABLED=1`	03/16/2023 1:35 AM UTC
5326e4e1845e55214886abcde61f4abd34819d2209a0bb3c0a278a9d5ebecc28	RUN	`PYVER=3.8 pip install --no-cache-dir git+https://github.com/NVIDIA/TransformerEngine.git@release_v${TRANSFORMER_ENGINE_VERSION}`	03/16/2023 1:35 AM UTC