Pointnet++ cuda 10.2 tensorflow 1.15 tf_ops 配置

找到libtensorflow_framework.so.1

miniconda下该文件目录

/home/jinwei/miniconda3/lib/python3.7/site-packages/tensorflow_core/libtensorflow_framework.so.1

到该目录下使用软连接

ln -s libtensorflow_framework.so.1 libtensorflow_framework.so

配置三个.sh 文件

tf_sampling_compile.sh

#/bin/bash
#/usr/local/cuda-8.0/bin/nvcc tf_sampling_g.cu -o tf_sampling_g.cu.o -c -O2 -DGOOGLE_CUDA=1 -x cu -Xcompiler -fPIC

/usr/local/cuda/bin/nvcc tf_sampling_g.cu -o tf_sampling_g.cu.o -c -O2 -DGOOGLE_CUDA=1 -x cu -Xcompiler -fPIC

TF_INC=$(python -c 'import tensorflow as tf; print(tf.sysconfig.get_include())')
TF_LIB=$(python -c 'import tensorflow as tf; print(tf.sysconfig.get_lib())')

#TF_INC=/home/vcortex/anaconda3/envs/tensorflow/lib/python3.7/site-packages/tensorflow_core/include
#TF_LIB=/home/vcortex/anaconda3/envs/ten