First, install anaconda
Here are installing Anaconda3-2019.03-Windows-x86_64.exe (python3.7 version)
Second, check the installation environment
1, if the anaconda installation checked configuration environment variable, direct cmd; if not checked, enter the Anaconda Prompt window (recommended)
2, detection anaconda environment is installed successfully:
conda --version
3, which detects the current installation environment:
conda info --envs
Third, the installation tensorflow in the anaconda
1, create tensorflow environment, install python3.7:
conda create --name tensorflow python=3.7
Note: tensorflow here is their definition of the name of the environment, corresponding to the anaconda installation directory envs folder.
Note: Due to Tsinghua University and other units anaconda mirror sites without obtaining authorization led to the closure, there is no replacement anaconda repository mirroring, you can use the default.
Note: If the download is interrupted due to network instability, try again.
2, the environment detection tensorflow Anaconda added to the inside:
conda info --envs
3, activation tensorflow environment (important, and then activate the next step):
source activate tensorflow
Note: This corresponds to the name of the environment tensorflow Step 1.
Note: Deactivation with conda deactivate
4, installation tensorflow:
pip install --upgrade --ignore-installed tensorflow
Note: Changing the pip download the source can improve download speeds. In C: \ Users \ username \ pip New pip.ini file folder:
[global]
index-url = https://mirrors.aliyun.com/pypi/simple/
[install]
trusted-host = mirrors.aliyun.com
Fourth, verify that the normal installation
1, in tensorflow environmental inputs: Python
2, the progressive input:
import tensorflow as tf
hello = tf.constant('hello,tf')
sess = tf.Session()
print(sess.run(hello))
3, print: b'hello, tf ', is the normal installation.