TensorRT | NVIDIA NGC

NVIDIA

TensorRT

Container

NVIDIA

TensorRT

NVIDIA TensorRT is a C++ library that facilitates high-performance inference on NVIDIA graphics processing units (GPUs). TensorRT takes a trained network and produces a highly optimized runtime engine that performs inference for that network.

NVIDIA AI Enterprise Supported

Layer	Label		Created
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	LABEL	`com.nvidia.build.ref=8d02910c12202507ed2952626547133155617a87`	06/03/2021 5:18 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	ARG	`NVIDIA_BUILD_REF`	06/03/2021 5:18 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	LABEL	`com.nvidia.build.id=23750633`	06/03/2021 5:18 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	ENV	`NVIDIA_BUILD_ID=23750633`	06/03/2021 5:18 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	ARG	`NVIDIA_BUILD_ID`	06/03/2021 5:18 PM UTC
sha256:9e79ee6ba22efaf9dd3cd3c99e4b344057f5b12cec1e0c39ee6d3466b76ac9cd	COPY	`NVIDIA_Deep_Learning_Container_License.pdf /workspace/`	06/03/2021 5:18 PM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4	ENTRYPOINT	`/usr/local/bin/nvidia_entrypoint.sh`	06/03/2021 5:18 PM UTC
sha256:e6cc5b05ccf7bf9b41e75c3a49515b61b4952e618bc954f2093192195c53bd80	COPY	`nvidia_entrypoint.sh /usr/local/bin`	06/03/2021 5:18 PM UTC
sha256:46485ae88ea336ba4f01b3652b404fe64eb4b4a251ec7bf1a2a492fad053dbf2	RUN	`NVIDIA_TENSORRT_VERSION=21.06 TRTOSS_VERSION=master 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`	06/03/2021 5:18 PM UTC
sha256:33fcf5b38ddabbd0d6f7c006e8e782772edae5868a8e6b4ccdcbcb4f113eb39d	COPY	`README.md README.md`	06/03/2021 5:18 PM UTC