Most things don’t come pre-installed for machine learning on windows laptops. In this guide we are going to go over some of the essential things you need to get started.
As a side note, you might also be interested in learning what we think is the best laptop for machine learning.
For machine learning most people use Python. There are other languages that are used, we go over some of them here. Despite there being other options, the easiest and most popular language is python.
Because of that, the first thing you should do to get your laptop set up for machine learning is to download Python for Windows. I recommend that you go for Python v3.6.
You can find the download links for Python here. There is going to be a lot of different option. You don’t have to use 3.6, but I picked v3.6.7.
Once the .exe has finished downloading launch the installer. Make sure you check the box that says “Add Python 3.x to PATH”. That way you are able to use the Python command from the terminal without having to do it yourself later. If you don’t click it then just search on Google for how to add it to the path yourself.
After you are done installing go to the command line and type “python –version”, and you should see something like this:
One of the most popular package managers on windows is Anaconda. You will see references too it all over the web, especially if you are looking at Python learning resources.
I recommend that you install Anaconda as your package manager.
Simply follow the installation guide for Anaconda on Windows that can be found here.
After you are done installing Anaconda you can type “anaconda –version” in the terminal. That should give you something like this:
If it says it is not a recognized command, first try to open a new command prompt if you already had one open. If it still says it then you mist likely did not check the button to add it to your path. You can either uninstall and do the installation over again, or you can look up how to add Anaconda to your path on Google.
Now we have both python and Anaconda. It is time to set up the packages that you are most likely to use. You can install all of these or a selection based on what you are interested in.
There might be other packages that you will need. But if you are following any tutorial online they will all say what to get and how to get them.
Machine learning is all about calculations of some sort. Arrays and calculations on those arrays are a big part of it. Numpy is a package that specializes in calculations on matrices and multi-dimensional arrays.
Install: pip3 install numpy
Scipy is another popular package for calculations. This package specializes in scientific calulations for things such as linear algebra, and image processing.
Install: pip3 install scipy
If you want to work with natural language processing then Nltk is for you.
Install: pip3 install nltk
Any data exploration that you do will involve plotting. Matplotlib is a standard tool for achieving many of the functionaries you will be looking for.
Install: pip3 install matplotlib
Pandas is an excellent tool for data analysis and working with data. Often used to read in data and perform queries on it.
Install: pip3 install pandas
This should be enough to get you started. You are know armed with the knowledge of how to install any package that you might need, just using pip3. You also have python installed so you can dive into learning more about machine learning and deep learning.