因為機器上裝的cuda tookit 是10.2版本 在測試tensorflow-gpu1.14.0的時候提示找不到庫文件
感謝博主https://blog.csdn.net/u014421914/article/details/103410545的提示，在tf路徑下創建這些缺失文件的軟連接如下
1 $PYTHON_PATH/site-packages/tensorflow$ ln -s '/usr/local/cuda/lib64/libcudart.so' ./libcudart.so.10.0 2 $PYTHON_PATH/site-packages/tensorflow$ ln -s '/usr/lib/x86_64-linux-gnu/libcublas.so' ./libcublas.so.10.0 3 $PYTHON_PATH/site-packages/tensorflow$ ln -s '/usr/local/cuda/targets/x86_64-linux/lib/libcufft.so' ./libcufft.so.10.0 4 $PYTHON_PATH/site-packages/tensorflow$ ln -s '/usr/local/cuda/targets/x86_64-linux/lib/libcurand.so' ./libcurand.so 5 $PYTHON_PATH/site-packages/tensorflow$ ln -s '/usr/local/cuda/targets/x86_64-linux/lib/libcurand.so' ./libcurand.so.10.0 6 $PYTHON_PATH/site-packages/tensorflow$ ln -s '/usr/local/cuda/targets/x86_64-linux/lib/libcusolver.so' ./libcusolver.so.10.0 7 $PYTHON_PATH/site-packages/tensorflow$ ln -s '/usr/local/cuda/targets/x86_64-linux/lib/libcusparse.so' ./libcusparse.so.10.0
其中PYTHON_PATH是自己python安裝目錄的路徑
附測試tensorflow GPU是否可用的方法
1 import tensorflow as tf 2 print(tf.test.is_gpu_available()) 3 print(tf.test.is_built_with_cuda())