Ubuntu18.04环境下使用Anaconda搭建Tensorflow

下载Anaconda

打开Anaconda下载地址，然后下载最新的Anaconda。
打开终端，进入Anaconda下载包所在的文件夹，运行bash Anaconda3-2020.02-Linux-x86_64.sh，一路y即可。
安装完成后运行conda --version检测是否安装成功。

搭建Tensorflow

在终端中输入conda create -n tensorflow python=3.7，一路y即可。
在终端中输入conda activate tensorflow 即可激活Tensorflow环境，conda deactivate即可退出Tensorflow环境。
在终端中输入人pip install --ignore-installed --upgrade tensorflow进行安装/升级。

此时，Tensorflow环境已经搭建完毕。

进行测试

在终端中输入touch test.py创建文件，修改文件内容为

import tensorflow as tf

import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

if __name__=='__main__':
    g = tf.Graph()
    # add ops to the user created graph
    with g.as_default():
        hello = tf.constant('Hello Tensorflow')
        sess = tf.compat.v1.Session()
        print(sess.run(hello))

在终端中输入python test.py，结果如下

在Pycharm中整合Tensorflow环境

file -> new project
选择Existing interpreter -> Conda Environment -> Ok -> Create即可。

进行测试

创建test.py，输入如下代码：

import tensorflow as tf

import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

if __name__=='__main__':
    g = tf.Graph()
    # add ops to the user created graph
    with g.as_default():
        hello = tf.constant('Hello Tensorflow')
        sess = tf.compat.v1.Session()
        print(sess.run(hello))

运行结果如下：

通常会弹出一些提示信息

2020-04-20 17:43:50.307721: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer.so.6'; dlerror: libnvinfer.so.6: cannot open shared object file: No such file or directory
2020-04-20 17:43:50.307796: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer_plugin.so.6'; dlerror: libnvinfer_plugin.so.6: cannot open shared object file: No such file or directory
2020-04-20 17:43:50.307805: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:30] Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.

这些信息时提示当前系统没有安装TensorRT相关的内容，如果不需要GPU支持，直接忽略即可，解决这些Warning的方法：在终端中运行如下代码

# Add NVIDIA package repositories
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/cuda-repo-ubuntu1804_10.1.243-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu1804_10.1.243-1_amd64.deb
sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub
sudo apt-get update
wget http://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64/nvidia-machine-learning-repo-ubuntu1804_1.0.0-1_amd64.deb
sudo apt install ./nvidia-machine-learning-repo-ubuntu1804_1.0.0-1_amd64.deb
sudo apt-get update

# Install NVIDIA driver
sudo apt-get install --no-install-recommends nvidia-driver-430
# Reboot. Check that GPUs are visible using the command: nvidia-smi

# Install development and runtime libraries (~4GB)
sudo apt-get install --no-install-recommends 
    cuda-10-1 
    libcudnn7=7.6.4.38-1+cuda10.1  
    libcudnn7-dev=7.6.4.38-1+cuda10.1


# Install TensorRT. Requires that libcudnn7 is installed above.
sudo apt-get install -y --no-install-recommends libnvinfer6=6.0.1-1+cuda10.1 
    libnvinfer-dev=6.0.1-1+cuda10.1 
    libnvinfer-plugin6=6.0.1-1+cuda10.1

结果如下：