 So we have seen how to represent one variable with the histogram. We have seen how to represent a variable in a category or two categories with the cat plot, box plot, and so on. So first now, of course, a lot of time we want to show the relationship between two numerical variables. So for that, okay, I just here reset the default theme because we had changed it before. We have this cat plot, okay, quite classical one. We just show dots of points. And so for this, you give your X and your Y. It's always kind of the same syntax. You give your data, your X, your Y, and you get shown whatever you have there. All right. And then you have the same hue argument. You can also change the color, change the palette as we have done before. Okay, so there's not much new. Now, one thing that we may want to do, for example, is to say, okay, this is fair, you see that there is a great concentration there at the beginning and a few points which are high. So maybe we want to log transform the data. And for that, it's not too hard. So I can just do the same thing as before. Grab whatever was created by the scatter plot in an X. And then X dot sets X scale equal log sets the X to a log scale. All right. Then, you know, we have some very small numbers and with logs, sometimes small numbers can skew things up a bit. Especially since I think we have a couple of zeros there. And so I manually set the limit of the axis. So set X limit between three and 1000 and two kind of show this. And that's what we get there. So that we see now, okay, I have my scatter plot and I'm beginning to tune it a little. And now, of course, maybe you want to represent your categories. So what you can do is you can have one category, the passenger class you as color. So that's the hue argument. And here I set up a custom palette, tomato till mustard. Yeah, four of my three colors. And then style is a little bit like hue, but it doesn't act on colors. It acts on the style of the points. And so you have here the sex being represented as round and crosses, okay, for male and female there. So we have now one level also of information that we give there. So that's one way of doing that. I think it's quite useful. Personally, though, I don't really like symbols too much. I find them not so always so informative. So I will just show you one other way that I do things. And it's by just using more colors. And I use what we called paired color palettes. No, that's this paired color palettes. Okay, so there what I do is that I concatenate the class and sex. Okay, together I created like one color, if you will, with all six possible combinations between all there. I use that in my hue. I give here the argument hue order, which will sort them, which will like, let me define precisely in what order I want these categories to come by because I would like them here. By sorting them, I make sure that they appear in a sense, in an order that kind of makes sense, right? And I don't mix like the sex and the classes and so on and so forth. And then I use the paired palette. So again, I've given you a little link there to see all the palettes that exist. And this one have like several views. And each time you have like a solid dark and light hue and they come in pairs, which works fairly well for me because each time I have pair, male, female, male, female, male, female, right? And that gives me this, which is another, let's say, flavor for the plot that we have just done above. Right. All good so far. Yes. All right. So then as you can see now, we have kind of the main recipe. It's just that now we have another kind of plot. Okay. It's the scatter plot, but then the way that we enter in this plot and that we play around with this plot is always kind of the same. We have hue and we have plenty of little arguments that we play around with and we tweak and we change stuff until we have something visually pleasing for us. Now there are two other kind of plots that I want to show you because sometimes, you know, this is nice, but sometimes you don't want to show just points. So the first is the KDE plot. So we've done that already for just a single line. But if you give both an X and a Y to the KDE plot, it will do a 2D dimension, like a 2D density plot. So you can see here density of three categories of three iris species. This is with the iris dataset there, depending on simple width and simple length. And it's just the function that we have used before, okay, with just the X and Y. And here's this little threshold there placed with the number, basically the difference that you are expecting between the lines there. So if you tweak this threshold argument, you will get either less or more lines. Okay, so that's one of the little style. And then also, sometimes you have, of course, data which are organized in time or organized, you know, such that you want to draw lines for them. And for that, of course, it's line plot. Okay, it's actually, you know, you want to show a scatter plot then this function scatter plot, you want to show lines that it's line plot. And then the way that it works is also fairly similar to what we've seen before. You specify what is the X, what is the Y. And eventually what is the hue, so what should be the color and what should be the style. So there now it's not around and cross, but it's just solid line or dash line or dot line or something like this. And the way that line plot is worked, it creates little error bars around the line, which is by default a confidence in a 95% confidence interval. But there are plenty of argument to make it standard deviation, something custom, no lines or show everything on so and so forth. Right, there is quite a few number of options there for you. Right, so that's just also to show you a small example of what we do with this simple one function code there. Okay, and that was about what I wanted to show to you. This one, I will launch it right now, it takes a bit of time, but for that exploration, it's quite of nice to do sometimes what we call a pair plot. And that will just create a visualization. Let's hope it doesn't take too long because it has to show to you plenty of data there. So that will just show a number of variable and their relationship to one another. Does it want to plot? Looks like it doesn't want to plot. So I will sort that out later on, but I don't want to dawdle too much. And so last thing I want to show to you is how to write these plots to disk. It's actually super simple, right? So you create your plots. At some point, you're like, ah, I'm super happy. I want to share that to a colleague, right? One way of doing that is to just, you can actually click and drag or just right click and save this image as something, right? But sometimes with code, it's a bit nicer. You can just create your plot. Okay. It's with SNS.catplot then you catch it as a variable and then you just call this dot save fig and then whatever you give there, it will be saved. If you output something dot PNG, it will be a PNG format. If you have something that PDF, it would be a PDF format. It's really as simple as that. It's actually relatively smart. And of course, if you want to have fine control over your DPI, number of pixel, height, width, and so on and so forth, you can see that there are some of these arguments there to save fig function. It's actually super, super simple. To save figures. There we go. So I think this is one of that one. And now that should work. Does. Okay. Sorry about that one. And then you see that I've created this output dot PNG file. There. And if I do that one now, it should work as well. Of course, it takes a bit of time to compute everything. And now you can save it. I have also this file that has been created there. All right. So, all right. There you go. And that takes us to the entire material. So sorry. I went a little bit over the hour there, but I was, I hope that it was worth it for you. Sorry. There was this little mess up with the, the effraction data frame. So there you are. So this is, we have shown you plenty of ways to do that analysis in Python to get your table in modified play around with it and then show you a few tricks to start plotting. Of course, we have only scratched the surface there. There is much more to go on. I have, we have left plenty of links throughout the notebooks for so that you can go and see more materials. I've shown you also the example guide. Go and see more materials. I've shown you also the example galleries and so on so forth. So don't hesitate to go there to play around with that to experiment. And you should.