From d699805e24a2b8b9cd0b86a845c9f816101ecf6d Mon Sep 17 00:00:00 2001 From: George Wu Date: Wed, 30 Mar 2022 14:07:06 -0700 Subject: [PATCH] update TensorRT EP docs for ORT 1.11 (#11055) --- docs/build/eps.md | 10 +++++----- docs/execution-providers/TensorRT-ExecutionProvider.md | 3 ++- 2 files changed, 7 insertions(+), 6 deletions(-) diff --git a/docs/build/eps.md b/docs/build/eps.md index e88828f490..29aa1afaab 100644 --- a/docs/build/eps.md +++ b/docs/build/eps.md @@ -99,7 +99,7 @@ See more information on the TensorRT Execution Provider [here](../execution-prov * The path to the CUDA `bin` directory must be added to the PATH environment variable so that `nvcc` is found. * The path to the cuDNN installation (path to folder that contains libcudnn.so) must be provided via the cuDNN_PATH environment variable, or `--cudnn_home` parameter. * Install [TensorRT](https://developer.nvidia.com/tensorrt) - * The TensorRT execution provider for ONNX Runtime is built and tested with TensorRT 8.0.3.4. + * The TensorRT execution provider for ONNX Runtime is built and tested with TensorRT 8.2.3.0. * To use different versions of TensorRT, prior to building, change the onnx-tensorrt submodule to a branch corresponding to the TensorRT version. e.g. To use TensorRT 7.2.x, * cd cmake/external/onnx-tensorrt * git remote update @@ -136,7 +136,7 @@ Dockerfile instructions are available [here](https://github.com/microsoft/onnxru ### Build Instructions {: .no_toc } -These instructions are for JetPack SDK 4.6. +These instructions are for JetPack SDK 4.6.1. 1. Clone the ONNX Runtime repo on the Jetson host @@ -159,7 +159,7 @@ These instructions are for JetPack SDK 4.6. export PATH="/usr/local/cuda/bin:${PATH}" ``` -3. Install the ONNX Runtime build dependencies on the Jetpack 4.6 host: +3. Install the ONNX Runtime build dependencies on the Jetpack 4.6.1 host: ```bash sudo apt install -y --no-install-recommends \ @@ -169,7 +169,7 @@ These instructions are for JetPack SDK 4.6. 4. Cmake is needed to build ONNX Runtime. Because the minimum required version is 3.18, it is necessary to build CMake from source. Download Unix/Linux sources from https://cmake.org/download/ - and follow https://cmake.org/install/ to build from source. Version 3.21.1 has been tested on Jetson. + and follow https://cmake.org/install/ to build from source. Version 3.23.0 has been tested on Jetson. 5. Build the ONNX Runtime Python wheel: @@ -178,7 +178,7 @@ These instructions are for JetPack SDK 4.6. --use_cuda --cuda_home /usr/local/cuda --cudnn_home /usr/lib/aarch64-linux-gnu ``` - Note: You may optionally build with experimental TensorRT support. + Note: You may optionally build with TensorRT support. ```bash ./build.sh --config Release --update --build --parallel --build_wheel \ diff --git a/docs/execution-providers/TensorRT-ExecutionProvider.md b/docs/execution-providers/TensorRT-ExecutionProvider.md index 9777a81e84..bd78de5a09 100644 --- a/docs/execution-providers/TensorRT-ExecutionProvider.md +++ b/docs/execution-providers/TensorRT-ExecutionProvider.md @@ -27,6 +27,7 @@ Pre-built packages and Docker images are available for Jetpack in the [Jetson Zo |ONNX Runtime|TensorRT|CUDA| |---|---|---| |master|8.2|11.4| +|1.11|8.2|11.4| |1.10|8.0|11.4| |1.9|8.0|11.4| |1.7-1.8|7.2|11.0.3| @@ -40,7 +41,7 @@ For more details on CUDA/cuDNN versions, please see [CUDA EP requirements](./CUD See [Build instructions](../build/eps.md#tensorrt). -The TensorRT execution provider for ONNX Runtime is built and tested with TensorRT 8.0.3.4. +The TensorRT execution provider for ONNX Runtime is built and tested with TensorRT 8.2.3.0. ## Usage ### C/C++