 I'm building this tool called Plotly. You can visit us online at plot.ly. And we're also in the Chrome Store at Plotly. Plotly is a tool that makes it easier for people to analyze and visualize data collaboratively. You have already installed Plotly with Pip and import it. And you can sign up to our service with the API. So let's just make a new account called Montreal Python and use a fake email address called Montreal Python at Shark Lasers. And Plotly returns with an API key, a temporary password. And now we can get going. So let's initialize a Plotly object with our new username and our newly assigned API key. And let's make a plot. So we have exposed this function called iPlot for IPython. iPlot. And we could just do simple arrays. And we have an interactive graph in IPython. And this is being served from Plotly servers. And it is being saved in your newly created Plotly account. So we can do more complicated things. Pretty much all scientific plots are available. So we could, say, make a box plot. Let's import numpy and make a box plot. We can fill it in with some normally distributed random points. And we have a box plot that's interactive. And the plotting syntax for Plotly is an array of dictionaries. So because we're Pythonistas, we'll do a little list comprehension and make 20 box plots and plot them. So we're going to make an array of box plots in iPython with one line. So as I said, these plots are actually saved in the cloud in your Plotly account. So you can click this link and view them online. This is our web tool. I can sign in with our new account. It's just Montreal Python. And we generated a password. And in Plotly is a GUI for styling these graphs. So we can remove the legend and interact with this without using any more code. And we can change the colors, change the opacity. All these types of things that you might do in a script to style, you can now do in a GUI. So we can show all the individual points for each box plot, zoom in, pan around. So let's do some more interacting with our account. So say we have some more random data, and then we plot it and have a file name. So now we can view this in Plotly, edit it, and maybe we're interested in the distribution of these randomly distributed values. In Plotly, we can change the type. So let's look at a histogram of our values. So that's kind of interesting. Now, all of our files are saved over here, and this file is called random. Now, I can interact with this plot, again, with IPython. So if I create a plot in the web browser or someone shares a plot with me in Plotly, I can add existing data. I can add new data to the plots that I make. So for example, I remember from probability, say, 100, the value of the equation for a bell curve. So first, let's change the normalization. So x will go from negative 3 to 3. Let's do 100 points. And then y, the equation for a bell curve, I think it's 1 over square root of 2 times pi times e to the x squared over 2, I think. x, y, and we call this plot random. So file name is equal to random. And let's append data. So our file option will be append. And feud in Plotly. Change this to that. And now we have our plot of a bell curve and a histogram of some sample data. I can share this with my other graphing homies, like Matt Sunquist. Send him a message. And move things. Man, that's annoying. The button is hidden. Zoom out. Send. Zoom in. And now he has access to this plot. He can make comments if he wants to. Hopefully he's online tonight. And we'll see this. And then I can share this plot with anyone through a link. And now they can view it. Or with all my graphing homies on Facebook. And then they can check it out. And they can view the plot. And they can save it to their own Plotly account. And they can edit it, change the colors. And they can even view the data behind the plot. So this is Plotly. We have been around for about a year now. It's being built by myself, my brother, a few friends in the mile end. We have an API for Python. But we've built the same sort of functionality for MATLAB, Julia, R, Arduino, Pearl. And we're trying to make it easier for people to share and make graphs on the internet. Right now, if you were to make a histogram on the web, you wouldn't even think about doing it. You'd think about maybe doing it in Excel or a Matplot lib, taking a screenshot, uploading it to Dropbox, and then sending it out. We want to make that a lot simpler for people. You can upload data. You don't even need to code. You can just upload CSVs or Excel spreadsheets, and we'll parse it. And you can graph it from within our online tool. And so we've given this tool out as a beta so far for about six months. And we've seen people make these incredible graphs like this, the height and weight of NFL players. This graph was featured today on The Washington Post, embedded in 76,000 people saw it. More people making sports box plots of distributions of their favorite players, fully interactive graphs that you can save and edit and view the data behind. People are streaming data with Arduino, Raspberry Pi, Nest Thermostats, and sending it online. And apparently sharing it with 9,000 people. So this dude's temperature at his house on three different levels. Some person yesterday used our Python API and integrated with a Raspberry Pi and has a temperature controller and is sending it out to 800 of their friends. So we're really excited about this because we're taking, you know, we want the data to be free and we want it to be online. And we want people to access it easier and visualize it easier and index it. And we want Plotly to be sort of the place where if you want to see a graph of something, you would go to Plotly. Or if you want to see some data of some statistics, you would think about going to Plotly. That's what we're really excited about. And the first step of that was making this tool to make it easier to make graphs and share them on the web. Any questions?