 So this is the actual production over a single day. But something that we'll often want to do is to plot averages or some other statistic. And so, in this last bit, we will go over how to plot the average hourly production. And to do that, I'm going to create another new variable called solar hour, which is equal to the date time, but just the hour value. And this is something that you can do with one of the benefits of date time objects is that you can use a special set of commands that extract bits of time or dates for you without having to do some sort of complicated read the string and figure out what that first number is. So now we can see the hour is over here. And so we can go ahead and plot this. Say solar. Again, do a G online. Yes. And here say x equals our and why is still produced kilowatt hours. And so this is showing the essentially the total kilowatt hours produced over each hour in a day across the entire year. So we haven't gotten to the average yet, which is our goal of doing that. But to do that, we need to use a special plot plotting tool within ggplot. So we still tell it which data to use. But now, instead of saying geome, we say stats summary. And then it still sort of works. So we still say x, we still have an aes statement in which we have our x and y that we've been using. But outside of that aes statement, we need to specify the geome. And so we've sort of just done things a little bit backwards. So if you remember with the bar plot, we specify geome bar and then stat inside that. But in order to work with the line plot, we need to specify stat outside and line inside. And then we also need to give it the function that we're applying to the y axis, which we are going to use num pi dot mean. And now we can see that this looks a lot more like we would expect. We can see the average kilowatt hour produced it's less than three across each hour of the day, given all of the data in 2021. So we can see that there was a spike just after six. And by and large, on average, three o'clock tends to be the best time to generate solar power here in central Pennsylvania. If we wanted to show a confidence interval on this. I'm just going to copy this and paste it down here. And a second stat summary, maintain the same aes values, but our geome then can be ribbon. And our function isn't just on the y axis now it's on the data. And it's mean confidence level bootstrapping, which we will get into in the next lesson what that actually means. But for now, this is the way that we can get a 95% confidence interval. And I'm also going to specify alpha, so that we can see the line through the ribbon. And so now we can see how this ribbon will follow the data. And so this is a way that you can work with time series and create line plots that have different functionality whether that's showing the actual data, or the average, or the average plus a confidence interval.