How to install a working Intel Data Science Environment with Graphviz

Question

I have come across the Intel DAAL and Intel MKL for data science and I wanted to explore it. So I thought of installing on my system and have a hands-on experience with it. I surfed in various websites to install my work environment in a particular location. I have been trying for quite a long time and failed. Can anyone explain to me how do I install?

Answer 1

You need to download the latest version of Anaconda IDE.

After downloading that, download and install the latest version of Graphviz installer. When I installed current version was 2.38.msi 

Follow these steps:

1. Install Graphviz MSI for all users
2. Navigate the environment variables as https://t.ly/Gz359
3. For the Graphvizprovide the new environment path, 2 paths need to be provided
4. Close all the command and Environment windows
5. Now open the new command window and try to test the Graphviz by using the command below:    

 c:\Users\MyDrive>dot -v

 You will be viewing the versions and other information. That means you going in the right way.

Then close the report using Ctrl-c, window and finally reboot the PC

1. Now Navigate Anaconda prompt as Administrator and stay in the environment 

    a. In Windows 10, Search for anaconda and select the anaconda command prompt: 

        i. Right-click on it and select to run it as Administrator

          b. Navigate to the root of the (base) environment: 

        cd\

    c. Get a current list of existing environments: 

        i. conda env list

    d. Remove any unwanted environments: 

        i. conda env remove -n OldenvironmentName

    e. Create new desired environment for Intel Data Science (ids) with the most current Conda libraries, and supported Python version 3.x:

        i. conda create -n ids python=3 numpy pandas seaborn matplotlib scikit-learn daal4py jupyterlab -y

    f. Activate the new environment:

        i. conda activate ids

    g. Install Graphviz with pip: 

        i. pip install graphviz

    h. Install python support for Graphviz: 

        i. conda install pydot python-graphviz -y

    i. Check that dot is accessible via cmd prompt: 

        i. dot -v

    j. Ctrl-C to close the report

   k. Register the following for intel DataScience enhancements:

        i. set USE_DAAL4PY_SKLEARN=YES

        ii. python -c "import sklearn"

    l. Reboot our PC

If you want to know more about the Data Science then do check out the following Data Science link which will help you in understanding Data Science from scratch