What do I need to know about the pandas index? (Part 2)

Key Takeaways

hello and welcome back to my Q&A video series on the pandas library in Python and today's video is actually a continuation from last time when we were talking about the pandas index So today we're going to talk about the series index and as well talk about alignment so if you haven't watched that first video I would go ahead and watch that now uh and then come on back here okay so we are going to start with the same data set we used last time so we'll import pandas as PD and then uh we're going to use this data set of uh alcohol consumption by continent or Sorry by country so we're going to read CSV and it's bit. Le SL drinks by country okay and we'll take a look at the head and uh just like last time we saw in the that a data frame always has an index right here okay and what I want to show you is that a uh series also has an index okay and it comes from the data frame so if for example I select the continent series and I say do head you will see 0 1 2 3 4 which is the index for the series which came from the data frame so the index is on the left and the values are on the right now let's pretend that uh we didn't use the default index for the data frame and instead we set something else as the index of the data frame so let's say for example uh drinks. set index and we're going to set the index as the country uh in place equals true and let's take another another look at it and we see that the country has been turned into the index okay so now what's going to happen when we select the continent series okay so drinks. continent. head and this time we're seeing the same thing as last time okay the index is on the left the values are on the right this is just a panda series the real contents is this but the index just came from the data frame and is attached to each row okay now you've actually seen series um many times before in these videos and uh they all include an index and you probably just didn't notice it or notice what it was so for example if I say drinks continent. value counts this is actually a series and as such uh and let's just copy and paste this uh as such it has for example an index and here it is and it has values okay now because it's a series that is output and not some special value counts object or something we can uh actually use the index to select values from the series what I mean is uh I can take this and I can refer to an index such as uh let's say Africa and I put it in Brackets and I'm saying from this series find the index Africa and and show me the value okay this is kind of like how we used Loke uh with the data frame to say oh I want to pull out the contents by referring to the index and the column name in that case here I just say what's the index I'm looking for it's Africa and it shows me the value and again this worked because this is a series object okay now next I want to talk about sorting okay so uh again pasting that and do you remember how you can sort the values of a series uh we saw that in a previous video and you just say do sort values and that sorts in ascending order so now you can see that the series is sorted okay but what if I want to sort the index itself say in a sense sending order I can actually use something called sort index oops and there you go it's now in ascending alphabetical order okay so in the last video we talked about three reasons the index exists and it was identification selection and alignment and so I want to show you what alignment means um and for that I need to create another series okay so I'm going to call it people and I'm going to say pd. series and just follow along with me here I'm going to pass in uh 3 million and a list of the numbers 3 million 85,000 okay and I'm going to pass I'm going to say index equals and pass a list uh Albania and Andora and uh I'm going to say name equals population okay so this is one way to create a panda series and let's just look at it uh so the population of Albania is approximately 3 million population of Andor is about 85,000 and this is how I constructed the series I gave it the values I gave it the index and I gave it a name now this is my tiny data set um and let's say I want to use this plus the drinks data set to calculate the total beer servings for each country and I'm going to do that by multiplying these numbers so the number of people times the data in the beer serving series which tells you the average per person so I'm going to take take the average per person multiply it by the number of people okay to see the total beer servings per year in that country so all I'm going to do is say drinks. beer servings which by the way I'll just remind you of what that looks like here's the drinks. beer servings I'm going to say that times people and when I run that check out what happens okay obviously for uh countries that were not represented in this series uh it's going to say well I can't do that math because I need I need the number of people so it marks it with not a number which is means a missing value however for the ones in which it does have the population amount it does the multiplication and here's the thing it aligned them by the series it didn't just take these two numbers and multiply them by the first two rows it found Albania 3 million and multiplied 3 million by the beer servings number for Albania Andor did the same thing okay so that's uh in summary alignment allows us to put data together and work with it together even if it does n uh if it's not exactly the same length as long as you tell it which rows correspond to which other rows so I took the Albania data here and the andur data here and it knew to do the math because they shared based upon this shared index okay so uh I want to uh end today with a bonus as always and uh the the question I would pose is how if I just wanted to take this People series and add it to the data frame how would I do that and I'm going to use a method called pd. concat short for concatenation now concat can be used to concatenate rows uh on top of other rows or columns next to other columns and the way we control that is with the a parameter so um uh I'm going to say uh I'm going to pass it the I'm going to pass it a list with the drinks object which is the data frame and the People series and I'll say axis equals 1 which means essentially put these objects side by side and when we do that and say. head check that out we've got a new column and even though it didn't have complete data it put the data in the right spots because of the series index and that is the beauty of the automatic alignment that pandas does using the index okay so that's it for today please click subscribe if you'd like to see more videos like this I'll be talking about the multi-index uh which I know some of you want to hear about in a future video uh let me know in the qu in the comment section below if you have any questions and that's it so um thanks again for joining me and I hope to see you soon