TensorFlow™ is a symbolic mathematical system based on dataflow programming that is widely used for programming implementations of various machine learning algorithms, formerly known as Google's neural network algorithm library DistBelief [1].
Tensorflow has a multi-tier architecture, can be deployed on all kinds of servers, PC terminals and web pages and supports high performance numerical computation on GPUs and TPUs, and is widely used in product development within Google and scientific research in various fields [1-2].
tensorflow -gpu installation
First, install Anoconda
1. official website download click me:
2. Installation
Click python 3.6 version to automatically download the x64 version, download it, and then install it.
As shown in the picture, after checking the box, all the way to next
3. Open the terminal
1) Enter conda -version to view the version
2) Configure the Python environment
I installed python 3.5, you can choose yourselves accordingly!
conda create –n tensorflow python=3.5
3) Activate python environment: activate tensorflow
You can exit the current environment: deactivate tensorflow
Getting to the point: installing tensorflow
**The traditional method is to use pip install, which leads to downloading NVIDIA's cuda support and cdnn library later** **The traditional method is to use pip install, which leads to downloading NVIDIA's cuda support and cdnn library later
**I present here a simple method that does not rely on pip***
In the terminal, type
conda install tensorflow-gpu
2. Look at the contents of the package
As shown in the picture, cuda 9.0 and cdnn 7.1.4 have been automatically installed without having to download them.
3. After installation
Open Anaconda Prompt Test Successful
summarize
The above is a small introduction to the tensorflow -gpu installation ,the easiest way (do not have to install their own cuda, cdnn), I hope to help you, if you have any questions please leave me a message, I will reply to you in a timely manner. I would also like to thank you very much for your support of my website!
If you find this article helpful, please feel free to reprint it, and please note the source, thank you!