Data analysts and scientists are tasked with extracting information and insights from huge datasets. This course introduces the Seaborn Python library helping engineers communicate information via its high-level and powerful visualization tools.
As deep learning approaches to machine learning rise in popularity, models are increasingly hard to understand and pick apart. Consequently, the need for sophisticated visualizations of the data going into the model is becoming more and more urgent and important.
In this course, Visualizing Statistical Data Using Seaborn, you will work with Seaborn which has powerful libraries to visualize and explore your data. Seaborn works closely with the PyData stack - it is built on top of Matplotlib and integrated with NumPy, Pandas, Statsmodels, and other Python libraries for data science
You will start off by visualizing univariate and bivariate distributions. You will get to build regression plots, KDE curves, and histograms to extract insights from data.
Next, you will use Seaborn to visualize pairwise relationships of high dimensionality using the FacetGrid and PairGrid.
Plot aesthetics, color, and style are important elements to making your visualizations memorable. Given this, you will study the color palettes available in Seaborn and see how you can manipulate specific plot elements in our graph.
At the end of this course you will be very comfortable using Seaborn libraries to build powerful, interesting and vivid visualizations - an important precursor to using data in machine learning. Software required: Seaborn 0.8, Python 3.x.
A problem solver at heart, Janani has a Masters degree from Stanford and worked for 7+ years at Google. She was one of the original engineers on Google Docs and holds 4 patents for its real-time collaborative editing framework.
Course Overview Hi! My name is Janani Ravi, and welcome to this course on Visualizing Statistical Data Using Seaborn. A little about myself. I have a master's degree in electrical engineering from Stanford and have worked at companies such as Microsoft, Google, and Flipkart. At Google, I was one of the first engineers working on real-time collaborative editing in Google Docs, and I hold four patents for its underlying technologies. I currently work on my own stuff at Loonycorn, a studio for high-quality video content. In this course, you will work with Seaborn, which has powerful libraries to visualize and explore your data. Seaborn works closely with the PyData stack. It is built on top of matplotlib and integrated with NumPy and StatsModels and other Python libraries used for data science. We start this course off by visualizing univariate and bivariate distributions. We'll build regression plots, KDE curves, and histograms to extract insights from data. We'll also study how we can detect correlations in our data using heat maps. We'll then use Seaborn to visualize pairwise relationships of high dimensionality using the facet grid and the pair grid. These are examples of trellis plots which are a precursor to building ML models. And we'll explore a bike sharing dataset which allows us to make decisions about the kind of machine learning model we want to build. Plots to fix color and style are important elements to making your visualizations memorable. We'll study the color palettes available in Seaborn and see how we can manipulate specific plot elements in our graphs. At the end of this course, you'll be very comfortable using Seaborn libraries to build powerful, interesting, and vivid visualizations, an important precursor to using data in machine learning.
Visualizing Relationships and Distributions in Seaborn Hi, and welcome to this course on Visualizing Statistical Data Using the Seaborn library in Python. We'll start off by visualizing relationships and distributions in Seaborn in this very first module. For any data analyst or data scientist, Seaborn is a very powerful tool in their tool clip. That's because Seaborn is a powerful visualization library which allows them to explore relationships that might exist in data and exploit these relationships when they build their machine learning models. If you've worked with data before, it's quite likely that you've used Matplotlib. Seaborn is very closely integrated with it. In fact, it's built on top of Matplotlib. In fact, Seaborn is very closely coupled with the PyData stack. The PyData stack is comprised of libraries for numerical and statistical analysis such as NumPy and pandas. An important Python library that we generally use for plotting figures and graphs is Matplotlib. Matplotlib seeks to make easy things easy and hard things possible. And Seaborn is a perfect complement to this. You can think of Seaborn as a higher-level abstraction which allows you to build very complex visualizations with very little code. Seaborn makes production ready plots.
Building Trellis Plots in Seaborn Hi, and welcome to this module on Building Trellis Plots in Seaborn. Trellis plots allow you to graph multivariate data, data with multiple variables across multiple plots which are displayed in a grid format. In this module, we'll work with two types of trellis plots. The first is the FacetGrid, which allows us to visualize relationships between multiple variables separately. FacetGrids are a way to view conditional relationships by setting up a matrix of plots. Every cell in a plot satisfies a certain condition, and you can view relationships between other data based on that condition. The PairGrid, on the other hand, is also a way to view relationships across multiple variables but in a pairwise manner. So you can specify a bunch of column pairs, and you can view all the relationships laid out in the form of a matrix.
Controlling Plot Aesthetics and Style in Seaborn Hi, and welcome to this module on Controlling Plot Aesthetics and Style in Seaborn. You have the visualization you need. How do you make it look good? That's what this module is all about. We'll start off by exploring the themes that are present in Seaborn which govern plot aesthetics. There are a number of built-in themes that you can use right out of the box. Seaborn also gives you a wide variety of color palettes to choose from. You can specify qualitative, sequential, and diverging color palettes. Each has its own use cases. You can also specify your own color palette based on your branding or design sense. In this module, we'll also study how we can override very granular style details in our plot to get the perfect look and feel.