Python for Machine Learning - An Introduction
Ever wanted to create your own Machine Learning model? Well, this is absolutely the best place to start. In this video series, you’ll learn all about the four Python libraries that are at the core of ML models – Numpy, Pandas, Matplotlib, and Seaborn. This video will walk you through the very basics of each of them and detail the scope for the next 10-15 videos in which we’ll be exploring how to code using these four Python libraries. After you finish viewing this video and before you jump into the first Numpy coding tutorial, be sure to install the libraries and packages that I’ve linked through below!
Python Download Instructions: https://www.python.org/downloads/
Numpy Installation Instructions: https://numpy.org/install/
Pandas Installation Instructions: https://pandas.pydata.org/pandas-docs...
Matplotlib Installation Instructions: https://matplotlib.org/3.3.3/users/in...
Seaborn Installation Instructions: https://seaborn.pydata.org/installing...
Anaconda Installation Instructions (Jupyter comes along with this): https://docs.anaconda.com/anaconda/in...
Welcome to the python programming for machine learning series. Python is extremely powerful, easy to use, versatile, and has great readability, meaning the syntax is not confusing or misleading, which makes it a great first programming language. And as a benefit to AI, it has several programming libraries linked to it that are very useful for Machine Learning and Data Science.
In this series, we’ll be looking at the 4 most common and important python libraries for ML: Numpy, Pandas, Matplotlib, and Seaborn. Now as a quick side note, this series won’t go into the very basics of the Python programming language itself – mostly basics of Python for these 4 libraries. That said, you won’t have to be an expert on Python to understand the content in these videos, but a little background knowledge on programming as a whole might be helpful.
First, we’ll be looking at the NumPy python library. Numpy is a python library that provides a multidimensional array object. Numpy provides several built in operations for arrays such as sorting, selecting, mathematical operations, etc. Now in theory, numpy arrays look quite similar to python lists. But Numpy arrays are actually much more effective in terms of size, performance, and functionality. They take up less space in the memory of the machine, they have faster computational abilities than python lists, and they have more optimized and advanced functions.
Pandas is a powerful library that allows you to analyze data with Python. It takes data, for example in a CSV file, and creates a dataframe, basically a python object with rows and columns that looks like a table or an excel spreadsheet. This is much more efficient than data analysis using just Python lists/dictionaries/etc.
Matplotlib is a Python library for plotting data and creating data visualizations such as line graphs, bar graphs, scatter plots, histograms, 3D plots, pie charts, etc. Plots help us understand trends, patterns, and correlations between the variables we are working with. And seaborn, which is the last library we’ll be exploring, is an basically extension of matplotlib as it also helps with data visualization but has some more advanced features (more plot options, better personalization (fonts, color, etc.), and advanced output). Now for all four of these libraries, you’ll have to install some packages onto your machine. The installation instructions for each of the packages is linked in the description above.
You’ll also need to download a python IDE or a workspace on which you can type and run your code. I will be using Jupyter notebook that’s part of the Anaconda distribution for these videos, and I highly recommend you get Anaconda and use Jupyter because it is extremely easy to use and is highly efficient. If you already have a workspace, then you can skip this step, but if you don’t, then you can follow the instructions on the website linked in my description box below.
About Acadaimy
Acadaimy empowers individuals to rapidly learn and master the fascinating field of Artificial Intelligence through easy-to-consume bite-sized videos delivered on a weekly basis. Topics include artificial intelligence, machine learning, ai applications, supervised learning, unsupervised learning, pattern recognition, artificial general intelligence, deep learning, data science, big data and more. While most immediately suitable for beginners, Acadaimy's AI training and accompanying examples and projects aims to deliver value to anyone ultimately seeking expertise in artificial intelligence.
Subscribe to Acadaimy: https://www.acadaimy.com/youtube
