Fishのblog

Build on Linux#

1. Clone the repo-克隆项目或者手动去github项目下载压缩包并改名mmcv(注意版本，选择2.1.0或者2.2.0)

   1
   git clone https://github.com/open-mmlab/mmcv.git
   2
   cd mmcv

2. Install ninja and psutil to speed up the compilation

   1
   pip install -r requirements/optional.txt

3. Check the nvcc version (requires 9.2+. Skip if no GPU available.)

   1
   nvcc --version

   If the above command outputs the following message, it means that the nvcc setting is OK, otherwise you need to set CUDA_HOME.

   1
   nvcc: NVIDIA (R) Cuda compiler driver
   2
   Copyright (c) 2005-2020 NVIDIA Corporation
   3
   Built on Mon_Nov_30_19:08:53_PST_2020
   4
   Cuda compilation tools, release 11.2, V11.2.67
   5
   Build cuda_11.2.r11.2/compiler.29373293_0

   Note

   If you want to support ROCm, you can refer to AMD ROCm to install ROCm.

4. Check the gcc version (requires 5.4+)

   1
   gcc --version

5. Start building (takes 10+ min)

   1
   pip install -e .
   2
   如果这样后显示安装成功但是第6步验证时出现no module的报错，首先检查cuda环境变量是否完全正确，然后执行下面这行指令
   3
   MMCV_WITH_OPS=1 FORCE_CUDA=1 pip install . --no-build-isolation -v

6. Validate the installation

   1
   python .dev_scripts/check_installation.py