ubuntu16.04下tensorflow-0.10(cuda7.5)安装
来源:互联网 发布:长沙蓝狐网络官网 编辑:程序博客网 时间:2024/04/29 16:12
tensorflow-0.10(cuda7.5)安装
1.annaconda(python2.7)、cuda(7.5)、cudnn(5.1)参见另一篇文章《
ubuntu--pva-faster-rcnn》点击打开链接
2.tensorflow-0.10安装
Ubuntu/Linux 64-bit, CPU only, Python 2.7:
$sudo pip install --upgrade https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl
Ubuntu/Linux 64-bit, GPU enabled, Python 2.7. Requires CUDA toolkit 7.5 and CuDNN v5.1:
$sudo pip install --upgrade https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl
问题来啦:
1.could not find cudnnCreate in cudnn DSO; dlerror: /usr/local/lib/python2.7/dist-packages/tensorflow/python/_pywrap_tensorflow.so: undefined symbol: cudnnCreate
解决办法
export CUDA_HOME=/usr/local/cuda
export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda/lib64"
sudo ldconfig /usr/local/cuda/lib64
2.libjpeg.so.62: cannot open shared object file: No such file or directory
solution: sudo apt-get install libjpeg62-dev
- ubuntu16.04下tensorflow-0.10(cuda7.5)安装
- Ubuntu16.04安装CUDA7.5+Caffe+tensorflow
- ubuntu16.04安装 cuda7.5
- Ubuntu16.04下同时安装CUDA8.0和CUDA7.0
- Ubuntu16.04系统下CUDA7.5配置Caffe教程
- Ubuntu16.04系统下CUDA7.5配置Caffe教程
- ubuntu16.04+nvidia gt740m+cuda7.5+caffe安装、测试经历
- ubuntu16.04+nvidia gt740m+cuda7.5+caffe安装、测试经历
- ubuntu14.04下tensorflow环境配置(tensorflow0.12多种安装方式+cuda7.5升级8.0)
- ubuntu16.04 python3.5下安装tensorflow(cpu only)
- ubuntu16.04下安装TensorFlow(GPU加速)
- 在Ubuntu16.04下安装TensorFlow
- ubuntu16.04下安装tensorflow(二)
- ubuntu16.04 下caffe+tensorflow+GPU 安装
- tensorflow安装(ubuntu14.04+cuda7.5+anaconda+pycharm)
- Ubuntu16.04下安装tensorflow(Anaconda3+pycharm+tensorflow+CPU)
- Ubuntu16.04安装tensorflow
- Ubuntu16.04 Tensorflow安装
- 主席树学习--入门
- 区间树
- 深度学习原理部分数学公式
- 线程私有属性
- MySQL数据库基础
- ubuntu16.04下tensorflow-0.10(cuda7.5)安装
- org.jboss.tools
- Difference Row
- Scoket
- 2.9.PHP7.1 女神级教程-女神的私人信息 -【控制语句 循环】
- SpringMVC 基础教程 简单入门实例
- DBCP1.3连接泄露问题
- CodeForces 778B Bitwise Formula 解题报告
- jQuery dialog 学习笔记