Heatmaps

Loading...

來自 University of Michigan 的課程

Applied Plotting, Charting & Data Representation in Python

1706 個評分

University of Michigan

Applied Plotting, Charting & Data Representation in Python

1706 個評分

課程 2（共 5 門，Specialization Applied Data Science with Python）

This course will introduce the learner to information visualization basics, with a focus on reporting and charting using the matplotlib library. The course will start with a design and information literacy perspective, touching on what makes a good and bad visualization, and what statistical measures translate into in terms of visualizations. The second week will focus on the technology used to make visualizations in python, matplotlib, and introduce users to best practices when creating basic charts and how to realize design decisions in the framework. The third week will be a tutorial of functionality available in matplotlib, and demonstrate a variety of basic statistical charts helping learners to identify when a particular method is good for a particular problem. The course will end with a discussion of other forms of structuring and visualizing data. This course should be taken after Introduction to Data Science in Python and before the remainder of the Applied Data Science with Python courses: Applied Machine Learning in Python, Applied Text Mining in Python, and Applied Social Network Analysis in Python.

從本節課中

Module 3: Charting Fundamentals

In this module you will explore charting fundamentals. For this week’s assignment you will work to implement a new visualization technique based on academic research. This assignment is flexible and you can address it using a variety of difficulties - from an easy static image to an interactive chart where users can set ranges of values to be used.

Interactivity5:27

與講師見面

Christopher Brooks

Heatmaps are a way to visualize three-dimensional data and

to take advantage of spatial proximity of those dimensions.

Now, I honestly didn't think I would include heatmaps in this course.

I've seen enough bad heatmaps to last me a lifetime.

The problem is heatmaps are really quite powerful when you have the right data.

Weather data is a great example.

You have two dimensions, latitude and longitude,

and then we can overlay on top of this a third dimension, say,

temperature or rainfall amounts and use color to indicate its intensity.

Here's another good example.

This time from the Australian Transportation Safety Bureau.

And actually on the Wikipedia page for Heatmaps, it shows the path of an airline,

Malaysian Airlines 370, which went missing in 2014.

On top of the flight path is a heatmap showing the probability of the plane if it

had crashed.

This probability is also displayed as a continuous value, with cool colors,

blues, being used for low probabilities and

hot or warm colors, reds, being used for high probability.

So heatmaps aren't all bad.

But where they break down is when there's no continuous relationship

between dimensions.

Using a heatmap for categorical data, for instance, is just plain wrong.

It misleads the viewer into looking for patterns and

ordering through spatial proximity.

And any such patterns would be purely spurious.

Okay, now you've have the lecture, let's talk about the techniques.

In matplotlib, a heatmap is simply a two-dimensional histogram where the x and

the y values indicate potential points and

the color plotted is the frequency of the observation.

Earlier, we talked about different statistical distributions and

they work well for this demonstration.

In fact, let's redo our subplots where we showed a scatter plot and

two histograms, one for each dimension.

In that we define Y values to come from the normal distribution and

X values to come from a random distribution.

To plot this as a two-dimensional histogram,

we just pass it to the hist2d function.

See how the edge colors on the top and the bottom are very dark purple,

while there are greens and yellows in the middle horizon of the image.

Now we talked a bit about what happens when you increase or

decrease the number of bins with regular histograms.

But how do you think this changes when using heatmaps?

You see, as you increase the bins you get the same effect.

Every data point starts to fall into its own category.

So you get a more clear picture.

But everything eventually is just one color since every data point

ends up being unique.

Now a legend is important and heatmaps are no different.

When using the pyplot scripting layer,

this is as easy as calling the pyplot colorbar function.

Be warned though,

there's lots of accounting and bookkeeping happening there under the hood.

In fact, pyplot is finding the current rendered image,

generating the colorbar, resizing the existing axes and

using gridspec to add a new axes the size of the colorbar.

It's quite a bit of bookkeeping.

And that's all there is to this though.

Now that you've seen this data in two different ways, a heatmap,

where you have an X, a Y, and a color dimension, and

a scatterplot, where you have two histograms, one for each axis.

So which of these approaches do you see yourself using

when you have density graphs that you want to share with others?

探索我們的目錄

免費加入並獲得個性化推薦、更新和優惠。

Coursera 致力於普及全世界最好的教育，它與全球一流大學和機構合作提供在線課程。

© 2018 Coursera Inc. 保留所有權利。

通過 App Store 下載

通過 Google Play 獲取