---
title: 源码编译构建
description: "获取TensorFlow源码，构建Python wheel包前需要配置CUDA环境，然后安装构建产物，配置运行库路径。"
url: https://www.hikunpeng.com/document/detail/zh/kunpengai/ecosystemEnable/tensorFlow/kunpengtensorflow_16_0005.html
sourcePath: /source/zh/kunpengai/ecosystemEnable/tensorFlow/kunpengtensorflow_16_0005.html
indexId: e5bfea40261dc90be8a9875028aacc0a0184f84ccfd231dc91318c9780f1ec6c78
---
# 源码编译构建

获取TensorFlow源码，构建Python wheel包前需要配置CUDA环境，然后安装构建产物，配置运行库路径。

1. 获取TensorFlow 2.21.0源码。
  1 git clone --recursive --branch v2.21.0 https://github.com/tensorflow/tensorflow.git

2. 配置CUDA构建环境。
  1 2 3 4 5 6 7 cd tensorflow export PYTHON_BIN_PATH=$(command -v python3) export TF_NEED_CUDA=1 export TF_CUDA_COMPUTE_CAPABILITIES=8.0 export CUDA_HOME=/usr/local/cuda export PATH=${CUDA_HOME}/bin:${PATH} export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:${LD_LIBRARY_PATH:-}

3. 生成构建配置。
  1 python3 configure.py

4. 构建Python wheel包。
  1 bazelisk build --config=opt --config=cuda //tensorflow/tools/pip_package:wheel

5. 安装构建产物。
  1 python3 -m pip install "bazel-bin/tensorflow/tools/pip_package/wheel_house/tensorflow-*.whl[and-cuda]"

6. 配置NVIDIA pip包安装的CUDA运行库路径。
  1 2 3 4 5 6 7 8 9 10 11 12 export LD_LIBRARY_PATH=/usr/lib64:\ /usr/local/lib/python3.11/site-packages/nvidia/cublas/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/cuda_cupti/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/cuda_nvrtc/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/cuda_runtime/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/cudnn/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/cufft/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/curand/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/cusolver/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/cusparse/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/nccl/lib:\ /usr/local/lib/python3.11/site-packages/nvidia/nvjitlink/lib:${LD_LIBRARY_PATH:-}