From 92efd9cb7b768924bc3868e7fff81eb332e75f08 Mon Sep 17 00:00:00 2001 From: AllentDan <41138331+AllentDan@users.noreply.github.com> Date: Tue, 10 Jan 2023 20:03:32 +0800 Subject: [PATCH] [Doc] update build doc for jetsons (#1621) * update build doc for jetsons * set lint only true for docs * change base revision of circleci config Co-authored-by: RunningLeon --- .circleci/config.yml | 3 ++- docs/en/01-how-to-build/jetsons.md | 18 ++++++++++++++++-- docs/zh_cn/01-how-to-build/jetsons.md | 17 ++++++++++++++++- 3 files changed, 34 insertions(+), 4 deletions(-) diff --git a/.circleci/config.yml b/.circleci/config.yml index a7405f052..06cd0f68e 100644 --- a/.circleci/config.yml +++ b/.circleci/config.yml @@ -26,12 +26,13 @@ workflows: csrc/.* lint_only false demo/csrc/.* lint_only false docker/.* lint_only false + docs/.* lint_only true mmdeploy/.* lint_only false requirements/.* lint_only false tests/.* lint_only false third_party/.* lint_only false tools/.* lint_only false - base-revision: master + base-revision: dev-1.x # this is the path of the configuration we should trigger once # path filtering and pipeline parameter value updates are # complete. In this case, we are using the parent dynamic diff --git a/docs/en/01-how-to-build/jetsons.md b/docs/en/01-how-to-build/jetsons.md index 593cdba89..03ce9db7b 100644 --- a/docs/en/01-how-to-build/jetsons.md +++ b/docs/en/01-how-to-build/jetsons.md @@ -3,7 +3,6 @@ In this chapter, we introduce how to install MMDeploy on NVIDIA Jetson platforms, which we have verified on the following modules: - Jetson Nano -- Jetson Xavier NX - Jetson TX2 - Jetson AGX Xavier @@ -174,12 +173,27 @@ It takes about 1 hour 40 minutes to install MMCV on a Jetson Nano. So, please be #### Install ONNX +Don't install the latest ONNX. The recommended ONNX version is 1.10.0. + ```shell # Execute one of the following commands -pip install onnx +pip install onnx==1.10.0 conda install -c conda-forge onnx ``` +If the installation failed and showed the following error: + +``` +CMake Error at CMakeLists.txt:299 (message): + Protobuf compiler not found +``` + +Please install dependencies: + +```shell +sudo apt-get install protobuf-compiler libprotoc-dev +``` + #### Install h5py and pycuda Model Converter employs HDF5 to save the calibration data for TensorRT INT8 quantization and needs `pycuda` to copy device memory. diff --git a/docs/zh_cn/01-how-to-build/jetsons.md b/docs/zh_cn/01-how-to-build/jetsons.md index e55a890b9..c589ebe45 100644 --- a/docs/zh_cn/01-how-to-build/jetsons.md +++ b/docs/zh_cn/01-how-to-build/jetsons.md @@ -156,12 +156,27 @@ conda activate mmdeploy - 安装 ONNX + 不要安装最新的 ONNX，推荐的 ONNX 版本是 1.10.0。 + ```shell # 以下方式二选一 - python3 -m pip install onnx + python3 -m pip install onnx==1.10.0 conda install -c conda-forge onnx ``` + 如果安装时，出现如下的报错信息: + + ``` + CMake Error at CMakeLists.txt:299 (message): + Protobuf compiler not found + ``` + + 请根据如下指令，安装依赖项: + + ```shell + sudo apt-get install protobuf-compiler libprotoc-dev + ``` + - 安装 h5py 和 pycuda Model Converter 使用 HDF5 存储 TensorRT INT8 量化的校准数据；需要 pycuda 拷贝显存