diff --git a/model_examples/GameFormer/README.md b/model_examples/GameFormer/README.md index d4d1f0d1bc87ef924d17bd4058798b105c385eb2..0a2b8ecab315fb0ff3f1248c4b3289b40ddde8a2 100644 --- a/model_examples/GameFormer/README.md +++ b/model_examples/GameFormer/README.md @@ -89,7 +89,7 @@ ``` 对于arm64架构Linux系统，waymo官方并没有提供预先编译好whl包。为了方便用户使用，我们提供arm64系统编译的whl包，可以直接在华为云OBS上进行下载。 ``` - wget --no-check-certificate https://pytorch-package.obs.cn-north-4.myhuaweicloud.com/DrivingSDK/packages/waymo_open_dataset_tf_2.11.0-1.5.0-py3-none-any.whl + wget https://pytorch-package.obs.cn-north-4.myhuaweicloud.com/DrivingSDK/packages/waymo_open_dataset_tf_2.11.0-1.5.0-py3-none-any.whl pip install -r requirements_waymo.txt pip install waymo_open_dataset_tf_2.11.0-1.5.0-py3-none-any.whl ``` @@ -106,7 +106,7 @@ 4. 安装 tcmalloc 高效内存资源分配库 ``` mkdir gperftools && cd gperftools - wget --no-check-certificate https://github.com/gperftools/gperftools/releases/download/gperftools-2.16/gperftools-2.16.tar.gz + wget https://github.com/gperftools/gperftools/releases/download/gperftools-2.16/gperftools-2.16.tar.gz tar -zvxf gperftools-2.16.tar.gz cd gperftools-2.16 ./configure diff --git a/model_examples/HPTR/README.md b/model_examples/HPTR/README.md index b2eed92194f19fa96969465425b7a92b89e48171..8e3254268788d4f9d224d016aa974734bcb959a6 100644 --- a/model_examples/HPTR/README.md +++ b/model_examples/HPTR/README.md @@ -96,7 +96,7 @@ HPTR 是一种用于自动驾驶实时运动预测的层次化Transformer模型 ``` 对于 arm64 架构 Linux 系统，waymo 官方并没有提供预先编译好 whl 包。为了方便用户使用，我们提供 arm64 系统编译的 whl 包，可以直接在华为云 OBS 上进行下载： ``` - wget --no-check-certificate https://pytorch-package.obs.cn-north-4.myhuaweicloud.com/DrivingSDK/packages/waymo_open_dataset_tf_2.11.0-1.5.0-py3-none-any.whl + wget https://pytorch-package.obs.cn-north-4.myhuaweicloud.com/DrivingSDK/packages/waymo_open_dataset_tf_2.11.0-1.5.0-py3-none-any.whl pip install waymo_open_dataset_tf_2.11.0-1.5.0-py3-none-any.whl ``` diff --git a/model_examples/LaneSegNet/README.md b/model_examples/LaneSegNet/README.md index 4a1e96bfb39ff0b44153f41aae183bb7e3606cfc..ed91100a45440b944fba19f59ed4553c8c302586 100644 --- a/model_examples/LaneSegNet/README.md +++ b/model_examples/LaneSegNet/README.md @@ -119,9 +119,8 @@ code_path=DrivingSDK/model_examples/LaneSegNet - 安装毕昇编译器 - 将CANN包安装目录记为cann_root_dir，执行下列命令安装毕昇编译器。 + 将CANN包安装目录记为cann_root_dir，执行下列命令安装毕昇编译器，官网下载毕昇编译器4.1.0版本：https://www.hikunpeng.com/zh/developer/devkit/download/bishengcompiler。 ``` - wget https://kunpeng-repo.obs.cn-north-4.myhuaweicloud.com/BiSheng%20Enterprise/BiSheng%20Enterprise%20203.0.0/BiShengCompiler-4.1.0-aarch64-linux.tar.gz tar -xvf BiShengCompiler-4.1.0-aarch64-linux.tar.gz export PATH=$(pwd)/BiShengCompiler-4.1.0-aarch64-linux/bin:$PATH export LD_LIBRARY_PATH=$(pwd)/BiShengCompiler-4.1.0-aarch64-linux/lib:$LD_LIBRARY_PATH @@ -130,7 +129,7 @@ code_path=DrivingSDK/model_examples/LaneSegNet - 安装依赖，将安装mpdecimal依赖包的目录记为mpdecimal_install_path。 ``` - wget --no-check-certificate https://www.bytereef.org/software/mpdecimal/releases/mpdecimal-2.5.1.tar.gz + wget https://www.bytereef.org/software/mpdecimal/releases/mpdecimal-2.5.1.tar.gz tar -xvf mpdecimal-2.5.1.tar.gz cd mpdecimal-2.5.1 bash ./configure --prefix=mpdecimal_install_path diff --git a/model_examples/MapTR/README.md b/model_examples/MapTR/README.md index 5510d8a5f0b4722fba4d8fd7daf9a27895d0b3f4..d89446b3a9df9563dab000fafa7e3a2b0f780666 100644 --- a/model_examples/MapTR/README.md +++ b/model_examples/MapTR/README.md @@ -148,9 +148,8 @@ MapTR是一种高效的端到端Transformer模型，用于在线构建矢量化 - 安装毕昇编译器 - 将CANN包安装目录记为cann_root_dir，执行下列命令安装毕昇编译器。 + 将CANN包安装目录记为cann_root_dir，执行下列命令安装毕昇编译器，官网下载毕昇编译器4.1.0版本：https://www.hikunpeng.com/zh/developer/devkit/download/bishengcompiler。 ``` - wget https://kunpeng-repo.obs.cn-north-4.myhuaweicloud.com/BiSheng%20Enterprise/BiSheng%20Enterprise%20203.0.0/BiShengCompiler-4.1.0-aarch64-linux.tar.gz tar -xvf BiShengCompiler-4.1.0-aarch64-linux.tar.gz export PATH=$(pwd)/BiShengCompiler-4.1.0-aarch64-linux/bin:$PATH export LD_LIBRARY_PATH=$(pwd)/BiShengCompiler-4.1.0-aarch64-linux/lib:$LD_LIBRARY_PATH @@ -159,7 +158,7 @@ MapTR是一种高效的端到端Transformer模型，用于在线构建矢量化 - 安装依赖 ``` - wget --no-check-certificate https://www.bytereef.org/software/mpdecimal/releases/mpdecimal-2.5.1.tar.gz + wget https://www.bytereef.org/software/mpdecimal/releases/mpdecimal-2.5.1.tar.gz tar -xvf mpdecimal-2.5.1.tar.gz cd mpdecimal-2.5.1 bash ./configure --prefix=/path/to/install/mpdecimal diff --git a/model_examples/PivotNet/README.md b/model_examples/PivotNet/README.md index a5063f72141239b32cfdcfcf971711c6a2040e34..11fcb6bfac4516db0333347ad667238058b92be0 100644 --- a/model_examples/PivotNet/README.md +++ b/model_examples/PivotNet/README.md @@ -130,9 +130,8 @@ code_path=model_examples/PivotNet - 安装毕昇编译器 - 将CANN包安装目录记为cann_root_dir，执行下列命令安装毕昇编译器。 + 将CANN包安装目录记为cann_root_dir，执行下列命令安装毕昇编译器，官网下载毕昇编译器4.1.0版本：https://www.hikunpeng.com/zh/developer/devkit/download/bishengcompiler。 ``` - wget https://kunpeng-repo.obs.cn-north-4.myhuaweicloud.com/BiSheng%20Enterprise/BiSheng%20Enterprise%20203.0.0/BiShengCompiler-4.1.0-aarch64-linux.tar.gz tar -xvf BiShengCompiler-4.1.0-aarch64-linux.tar.gz export PATH=$(pwd)/BiShengCompiler-4.1.0-aarch64-linux/bin:$PATH export LD_LIBRARY_PATH=$(pwd)/BiShengCompiler-4.1.0-aarch64-linux/lib:$LD_LIBRARY_PATH @@ -141,7 +140,7 @@ code_path=model_examples/PivotNet - 安装依赖，将安装mpdecimal依赖包的目录记为mpdecimal_install_path。 ``` - wget --no-check-certificate https://www.bytereef.org/software/mpdecimal/releases/mpdecimal-2.5.1.tar.gz + wget https://www.bytereef.org/software/mpdecimal/releases/mpdecimal-2.5.1.tar.gz tar -xvf mpdecimal-2.5.1.tar.gz cd mpdecimal-2.5.1 bash ./configure --prefix=mpdecimal_install_path @@ -208,9 +207,9 @@ code_path=model_examples/PivotNet ``` cd /path/to/pivotnet cd assets/weights -wget --no-check-certificate https://github.com/wenjie710/PivotNet/releases/download/v1.0/efficientnet-b0-355c32eb.pth . -wget --no-check-certificate https://github.com/wenjie710/PivotNet/releases/download/v1.0/resnet50-0676ba61.pth . -wget --no-check-certificate https://github.com/wenjie710/PivotNet/releases/download/v1.0/upernet_swin_tiny_patch4_window7_512x512.pth . +wget https://github.com/wenjie710/PivotNet/releases/download/v1.0/efficientnet-b0-355c32eb.pth . +wget https://github.com/wenjie710/PivotNet/releases/download/v1.0/resnet50-0676ba61.pth . +wget https://github.com/wenjie710/PivotNet/releases/download/v1.0/upernet_swin_tiny_patch4_window7_512x512.pth . ``` - 生成模型训练数据 diff --git a/model_examples/QCNet/README.md b/model_examples/QCNet/README.md index 56f235489fe7aeda8c90e8cf10936ddc97fa2be1..9c9461779c47c936811a4e5ca82e2886183241a0 100644 --- a/model_examples/QCNet/README.md +++ b/model_examples/QCNet/README.md @@ -132,7 +132,7 @@ code_path=model_examples/QCNet 5. 安装 tcmalloc 高效内存资源分配库 ``` mkdir gperftools && cd gperftools - wget --no-check-certificate https://github.com/gperftools/gperftools/releases/download/gperftools-2.16/gperftools-2.16.tar.gz + wget https://github.com/gperftools/gperftools/releases/download/gperftools-2.16/gperftools-2.16.tar.gz tar -zvxf gperftools-2.16.tar.gz cd gperftools-2.16 ./configure diff --git a/model_examples/StreamPETR/README.md b/model_examples/StreamPETR/README.md index bf25316c72ca28ba967ec75622b97cf979cde2bf..793d576aeb73712cb9b5bd7c767d5f37e12b217f 100644 --- a/model_examples/StreamPETR/README.md +++ b/model_examples/StreamPETR/README.md @@ -226,10 +226,9 @@ chmod 777 StreamPETR/backfunc.sh 10. 编译优化 (1)python编译优化 -源码下载毕昇编译器： +官网下载毕昇编译器4.1.0版本：https://www.hikunpeng.com/zh/developer/devkit/download/bishengcompiler： ``` ulimit -n 4096 -wget --no-check-certificate https://kunpeng-repo.obs.cn-north-4.myhuaweicloud.com/BiSheng%20Enterprise/BiSheng%20Enterprise%20203.0.0/BiShengCompiler-4.1.0-aarch64-linux.tar.gz tar -xvf BiShengCompiler-4.1.0-aarch64-linux.tar.gz export PATH=$(pwd)/BiShengCompiler-4.1.0-aarch64-linux/bin:$PATH export LD_LIBRARY_PATH=$(pwd)/BiShengCompiler-4.1.0-aarch64-linux/lib:$LD_LIBRARY_PATH @@ -238,7 +237,7 @@ source {cann_root_dir}/set_env.sh 下载所需依赖： ``` -wget --no-check-certificate https://www.bytereef.org/software/mpdecimal/releases/mpdecimal-2.5.1.tar.gz +wget https://www.bytereef.org/software/mpdecimal/releases/mpdecimal-2.5.1.tar.gz tar -xvf mpdecimal-2.5.1.tar.gz cd mpdecimal-2.5.1 bash ./configure --prefix=/path/to/install/mpdecimal