Detailed explanation of winsorflow quantum installation process

The content of other blogs quoted in this blog will be marked with the links of the original blogger's articles .
Main reference ** windows System installation tensorflow quantum course **

1.Anaconda install

Anaconda Installation tutorial

• When installing , You can check all of the following options , There is no need to configure environment variables .
  

2.tensorflow2.1.0 install

Direct up link , Another blogger's tutorial , Write very well . It's just that you need to put tensorflow2.0.0 Change the version to the one we want to install tensorflow2.1.0 The version is fine .

tensorflow2.0.0 Installation
During the installation , Best network speed , Otherwise, there will be problems in downloading some files , Lead to in spyder Error occurs when running code

installed tensorflow-gpu=2.1.0 when , Some courses rely on lower or higher versions of packages , There will be conflicts , If you report an error
direct pip3 install joblib -i https://pypi.doubanio.com/simple Other dependency conflicts , Just put joblib Replace with dependency conflicts

3.tensorflow-quantum install

adopt cmd Order to enter windows Command line operation interface of the system , And then use conda activate XXX（ This is the second step to create tensorflow2.1.0 The name of the virtual environment ） Activate the virtual environment created in the second step , And then use ：

pip3 install --upgrade pip
pip3 install -U tensorflow-quantum
pip3 install -U tfq-nightly

4. Summarize the problems arising from the process operation

In the installation tensorflow when , Pay attention to tensorflow、CUDA、python Version issue for . The version correspondence of the three is attached below :
tensorflow、CUDA、python Version correspondence

problem 1：CUDA driver version is insufficient for CUDA runtime version
Which translates as CUDA The driver version of is the same as CUDA The runtime version of does not match ！
Details can be viewed CUDA driver version is insufficient for CUDA runtime version Problem solving
This is the kind of problem I have , After installation tensorflow-quantum after , stay spyder This problem occurs when you run code on , Finally, upgrade the computer CUDA Version solves the problem .
Follow the tutorials written by the following two bloggers :

Win10 install CUDA10 and cuDNN

CUDA Toolkit 10.0 Archive download

I just summarize the problems encountered in the learning process , I hope I can help people in need .