Microsoft's Deep Learning Framework (cntk), I have seen a framework with the simplest installation method. After 2.0, I started to support C,
Wiki: https://github.com/Microsoft/CNTK/wiki
Hi, are you a zombie like me? Previously, I tried to install mxnet and tensorflow. However, due to a short period of time, I often lost my interest in installing mxnet and tensorflow. Today I saw that Microsoft's cntk is ready for use. I installed mxnet at one time and passed the test.
My environment:
Windows 7X64
8 GB memory
Download link
Official installation instructions: https://github.com/Microsoft/CNTK/wiki/Setup-Windows-Binary-Script
After downloading the installation package with the above link, copy and paste the installation package:
C: \ local \ cntk
This directory needs to be used later
Directory after decompression
C: \ local \ cntk \ Scripts \ install \ windows
The following file is named install. bat. Then you can run the script to complete the installation process.
It will help you install the related class libraries and software packages required by the entire environment...
The following content references official instructions:
- VS2015 Runtime will be installed
- MSMPI will be installed
- Anaconda3 will be installed into the folder
C:\local\Anaconda3-4.1.1-Windows-x86_64
- A CNTK-PY35 environment will be created or updated in
C:\local\Anaconda3-4.1.1-Windows-x86_64\envs
- The CNTK Python module will be installed or updated in the created CNTK-PY35 environment
- A batch file will be created to activate the created Python environment and set the required environment variables
The third official step is to install and upgrade the video card driver. I skipped this step because my video card does not meet the requirements.
Step 4
First:
Run the following code to activate the CNTK environment.
cd C:\local\cntkcd scriptscntkpy35.bat
The next step is to run the framework according to the official instance code:
cd C:\local\cntk
cd Tutorials\HelloWorld-LogisticRegression cntk configFile=lr_bs.cntk makeMode=false command=Train
The final result is shown in:
I don't know what I did, but at least I have installed a deep learning environment on the tall, right?