Data Science & Machine Learning - 3.1 NumPy & Its Installation

Hi friends,

Welcome to another post under Data Science & Machine Learning tutorials. In the previous few posts, we reviewed basic Python syntax to refresh our memory with Python programming language. In this post, we'll see the NumPy library in Python. It is the fundamental library for scientific computing with Python. It supports various linear algebra operations and almost all of the libraries of PyData Ecosystem depend on NumPy which makes it even more important. It is incredibly fast and also supports tools for integrating C/C++ code. 

NumPy Installation

For installation of NumPy, it is recommended to use the Anaconda distribution of Python in order to get all the underlying dependencies to get installed on its own. You can see the installation of Anaconda distribution of Python here. Once you have that installed, you can install NumPy by running the following command in the command prompt:

conda install numpy

You can still install NumPy even if you don't have the Anaconda distribution of Python (not recommended) using the following command:

pip install numpy

We'll mostly be working on NumPy Arrays supported by NumPy library. NumPy Arrays support two ways of usage:

1. Vectors - These are 1-D NumPy arrays
2. Matrices - These are multi-dimensional NumPy arrays. 

However, we can still have matrices with just one row or just one column which is basically nothing but another representation of a vector. 

In the next post, we'll see about the NumPy Arrays and various data science libraries that operates on NumPy Arrays in detail.