From f693a6d723ef2766be36deb5e7987cd50159973a Mon Sep 17 00:00:00 2001 From: Rajeev Rao Date: Fri, 4 Sep 2020 13:43:45 -0700 Subject: [PATCH] README fixes Signed-off-by: Rajeev Rao --- README.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/README.md b/README.md index 7085ddee..501a92d8 100644 --- a/README.md +++ b/README.md @@ -87,7 +87,7 @@ NOTE: Along with the TensorRT OSS components, the following source packages will git clone -b master https://github.com/nvidia/TensorRT TensorRT cd TensorRT git submodule update --init --recursive - $Env:TRT_RELEASE_PATH = $(Get-Location) + $Env:TRT_SOURCE = $(Get-Location) ``` 2. #### Download the TensorRT binary release. @@ -147,7 +147,7 @@ NOTE: Along with the TensorRT OSS components, the following source packages will ```powershell cd ~\Downloads Expand-Archive .\TensorRT-7.1.3.4.Windows10.x86_64.cuda-11.0.cudnn8.0.zip - $Env:TRT_RELEASE_PATH = '$(Get-Location)\TensorRT-7.1.3.4' + $Env:TRT_RELEASE = '$(Get-Location)\TensorRT-7.1.3.4' $Env:PATH += 'C:\Program Files (x86)\Microsoft Visual Studio\2017\Professional\MSBuild\15.0\Bin\' ``` @@ -155,7 +155,7 @@ NOTE: Along with the TensorRT OSS components, the following source packages will **JetPack example** - Using the SDK manager, download the host componets of the PDK version or Jetpack specified in the name of the Dockerfile. To do this: + Using the SDK manager, download the host components of the PDK version or Jetpack specified in the name of the Dockerfile. To do this: 1. [**SDK Manager Step 01**] Log into the SDK manager 2. [**SDK Manager Step 02**] Select the correct platform and Target OS System (should be corresponding to the name of the Dockerfile you are building (e.g. Jetson AGX Xavier, `Linux Jetpack 4.4`), then click `Continue` 3. [**SDK Manager Step 03**] Under `Download & Install Options` make note of or change the download folder **and Select Download now. Install later.** then agree to the license terms and click `Continue` @@ -207,6 +207,7 @@ NOTE: Along with the TensorRT OSS components, the following source packages will > NOTE: To run TensorRT/CUDA programs in the build container, install [NVIDIA Docker support](#prerequisites). Docker versions < 19.03 require `nvidia-docker2` and `--runtime=nvidia` flag for docker run commands. On versions >= 19.03, you need the `nvidia-container-toolkit` package and `--gpus all` flag. + ## Building The TensorRT OSS Components * Generate Makefiles and build. @@ -294,10 +295,10 @@ NOTE: Along with the TensorRT OSS components, the following source packages will whl files for the TensorRT python API are in the `python` directory of the TensorRT release -*Example* install for python 3.6: +*Example* install for python3: ``` -pip install $TRT_RELEASE/python/tensorrt-7.1.3.4-cp36-none-linux_x86_64.whl +pip3 install $TRT_RELEASE/python/tensorrt-7.1.3.4-cp36-none-linux_x86_64.whl ``` ## Useful Resources @@ -313,5 +314,4 @@ pip install $TRT_RELEASE/python/tensorrt-7.1.3.4-cp36-none-linux_x86_64.whl ## Known Issues #### TensorRT 7.1 -* [demo/BERT](demo/BERT) has a known accuracy regression for Volta GPUs; F1 score dropped (from 90 in TensorRT 7.0) to 85. A fix is underway. * See [Release Notes](https://docs.nvidia.com/deeplearning/tensorrt/release-notes/tensorrt-7.html#rel_7-1-3).