Configuring your PC for A.I. (20 minutes)

 There are several options to start developing your own A.I. programs.  By far the easiest is to use Google Colab with a Chrome browser.  Checkout the Colab Overview if you want to explore that option.  Instead, you can opt to develop your A.I. projects directly on your own PC.  You can use any PC to get started but if you end up developing more advanced A.I. projects you will have to have a good GPU (Graphics Processing Unit). 

Step 1 – Download a Package Manager (Anaconda Individual Edition)

Think of a package manager as a large free app store.  It will make it easy to download and update all the different libraries you will need to code your first A.I. project.   All you need to do is to download and install the package manager (Anaconda Individual Edition in this case) on your PC.  After that, all the other software you will need will be downloaded and installed by the package manager itself.  Makes it really simple!

So, head over to Anaconda and install the Anaconda Individual Edition using the link below:

https://www.anaconda.com

Step 2 – Create a Virtual Environment

Think of a virtual environment as a small fenced in area on your computer where you can develop your code.  You can opt to create many of these but only one is needed to get started.  To create your virtual environment, launch the program Anaconda Prompt.   This will open up a new window which will allow you to type in commands to build your virtual environment.   Begin by typing:

conda create –n python3 python=3
conda activate python3

Voila, you have now created your first virtual environment, awesome!  You now have Python on your computer as well as a nice fenced in area called “python3” where you can load all your other software that you will need.

Step 3 – Download and Install A.I. Libraries

Next you will use the package manager to download the various libraries you will need to start coding your first A.I. project.   Type the following commands in the Anaconda Prompt command line:

conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch
conda install matplotlib jupyter git

The first line will install PyTorch which is a popular open-source deep learning library that will enable you to build neural networks.   The second line will install some graphing utilities and an ability for you to code your project using just a web browser.

Step 4 – Create a New Directory and Download a Sample Project

You should now have Python and all the initial A.I. libraries you will need to get started.  Next, create a new directory and download a sample starter project by typing: 

cd\
mkdir mycode
cd mycode
git clone https://github.com/LeakyAI/FirstNeuralNet

The last line will download some A.I. starter code that will help you build your first neural network.

Step 5 – Launch Jupyter Notebook

Jupyter Notebook is a program that enables you to build and execute code in your own web browser.  As a last step, launch Jupyter Notebook on your local machine by typing:

jupyter notebook

Step 6 – Execute your First A.I. Project

Congratulations!  Your P.C. is now ready for you to start coding your own A.I. projects!  You can start by loading the notebook FirstNeuralNetworkSolution.ipynb.  This notebook will show you how to build a neural network that can predict lemonade sales using a synthetic dataset.  You can learn more about that by taking the free tutorial “Project Lemonade:  Predict Sales Using a Neural Network (30 Minutes)”.

Next Steps

If you want to learn more about developing your own A.I. projects, take our 4-week Introduction to A.I. programming course!

Happy Learning!

leaky.ai team