Plotly 1.5

Loading...

En provenance du cours de Johns Hopkins University

Developing Data Products

1404 notes

Johns Hopkins University

Developing Data Products

1404 notes

Cours 9 sur 10 dans Specialization Data Science

A data product is the production output from a statistical analysis. Data products automate complex analysis tasks or use technology to expand the utility of a data informed model, algorithm or inference. This course covers the basics of creating data products using Shiny, R packages, and interactive graphics. The course will focus on the statistical fundamentals of creating a data product that can be used to tell a story about data to a mass audience.

À partir de la leçon

Shiny, GoogleVis, and Plotly

Now we can turn to the first substantive lessons. In this module, you'll learn how to develop basic applications and interactive graphics in shiny, compose interactive HTML graphics with GoogleVis, and prepare data visualizations with Plotly.

Rencontrer les enseignants

Brian Caffo, PhD
Professor, Biostatistics
Bloomberg School of Public Health
Jeff Leek, PhD
Associate Professor, Biostatistics
Bloomberg School of Public Health
Roger D. Peng, PhD
Associate Professor, Biostatistics
Bloomberg School of Public Health

So the next couple examples are easy enough that I'm not going to

do code demonstrations, I'm just going to show you the output.

A histogram is easy enough, you just do type = "histogram" and

you need one x variable.

Show it, it looks like this.

It'll actually show you the height of the bars as you move your mouse along.

And you can change the colors, do side by side histograms, and

a variety of other things in plot_ly, and they all kind of work in the obvious way.

So for boxplot here, plot_ly data frame is iris in this case,

y is petal length, and then color is species and type = "box".

So this is going to create separate boxplots, one for each species type.

Okay, and here it is, our three species.

When you hover over the box it gives you the values for the boxplot, so

this the 25th percentile, the 75th percentile, the median.

And then the whiskers extend out to 1.5 times

the distance between these two values.

That was Tukey's idea, that references back to the normal distribution,

the 1.5 times the inter-quartile range.

So here's your boxplot.

If you have an outlier like this, it'll actually show you the little point for

the outlier, okay?

So this can get a little crowded if it's all squashed together like that one.

But still, nice plot for almost no effort.

So a heatmap is basically just a graphical display of an image.

In this case, our image is going to be a matrix that we create.

So pretty much all images, all raster images, are just a matrix, right?

They're rows and columns.

And then they have the intensity value as the value for

each specific row and column combination.

So in this case,

let's just crate a simple matrix where we have 100 rows and 100 columns.

And we just fill it in with random normals, okay?

And that's what plot_ly's expecting for a heatmap.

Okay, the z is just going to be this matrix.

And then the type is going to be heatmap.

Okay, and then when it does that it plots an image.

A heat map is ultimately just an image, okay, just a raster image.

Okay, and the nice thing about the way plot_ly does it is,

as you move your mouse around it gives you the x and the y values, and

the z values, the intensity values.

And then the intensity values are given a color key here on the right,

which is quite nice so you don't have to set that up, okay?

So that is a pretty easy way to generate a heat map, and an interactive one at that.

So you could imagine taking that heat map, and what if you wanted to plot z, instead

of it as just color, what if you wanted to plot it pulled up into a third dimension?

Well, that would be plotting a surface.

So why don't we create this same dataset,

this time with different random normals, and then our z is the same thing.

Our type is now surface instead of heatmap, okay?

And so the intensity value is going to be the z value, and then the row and

column values, it's just going to assign integers for

those, are just going to be the x and y values.

And then when you do that, you get a nice 3D surface.

Now whenever you do a 3D surface like this, it's doing some smoothing, okay?

So you need to, this is just to show you how the commands work, but

of course if you want to do this for

real, you are going to need to actually understand what the smoothing is doing.

Explorer notre catalogue

Rejoignez-nous gratuitement et obtenez des recommendations, des mises à jour et des offres personnalisées.

Coursera propose un accès universel à la meilleure formation au monde, en partenariat avec des universités et des organisations du plus haut niveau, pour proposer des cours en ligne.

© 2018 Coursera Inc. Tous droits réservés.

Télécharger dans l'App Store

Disponible sur Google Play