Data Slicing in Python with Mito

Key Takeaways

hey this is jake from mido thanks for checking out this video i'm going to show you how you can really easily slice and dice and transform your data using the mito python package so the first thing we're going to do is run these two lines of code here so import miter sheet and my do sheet dot sheet and that's going to render our mito front end so the miter front end what it is is it's a spreadsheet front end for python inside the jupyter notebook so every edit we make in this front end here is going to generate the equivalent python below the first thing let's do is just get our data in so we can get data in one of two ways one is we can just search our local files here i'm going to decide this uh netflix csv file so with some data about the different things you can watch on netflix so we'll do that and we see when we do that it populates the miter sheet with our data from this netflix csv file and below it generates the code that's turned that csv into a data frame the other way we can get data into the tool is we can pass in a data frame directly so you can call in the miter sheet at any point your analysis if you're working with data frames above all you have to do is pass in the name of the data frame as an argument to this minor sheet.sheetcall and it'll populate the sheet before we move forward just ways you can actually analyze and understand your data using mido let me just show you how you can install it really quickly so all you have to do is these three commands here from our documentation we're going to install the mito installer then run the install command from within the installer and then just open jupyter lab and you're good to go so back in mido we have some data that is about the different titles the different things you can watch on netflix so there's tv shows and movies obviously let's look at some summary statistics really quickly for the breakdown between tv shows and movies just to understand our data better i'll go to the summary stats tab here we can see we have a good amount more movies we see the exact number there five three seven seven uh more movies and tv shows we can see and the other data point i'm interested in here is the rating so what these movies and tv shows are rated again i'm going to look at the summary stats for those here we can see tvma tv 14 and tvpg are the most common ratings and again we can see the exact values for each of them on those labels that pop up so what i want to do here now is just understand the relationship a bit better to really slice and dice my data in a way that gives me some insights so i'm going to do a pivot table which is a great way to do that i'll click pivot first thing i can do here is i'm going to add the ratings as my row and as the columns i'm not going to put the type so it's either going to be tv show or movie and as the value i'll just select ratings again and just put the aggregation type as count and what that does there is it populates this here so i'll close this pivot table now and now we've made this really great um editable updating pivot table inside my own so we can see for example the rating tv 14 the netflix profile has 1272 movies that have that rating and 659 tv shows that have that rating and again as i said before everything we do here is generating the equivalent code below so when i make this pivot table below it generates the code that is the that represents this exact same pivot table so if we were to just do this by hand it would generate this same pivot table here um and what i can do here is really nice about the generated code is i can use that carry forward in my analysis so this is all real code i can use so i can run this cell here and now this pivot table is called df2 as we can see here if i were to print out df2 right here we would see the same pivot table that we have in the cell above here another thing i might want to do to really understand my data and you know help transform and help decide how i want to analyze more is put a graph on top of this represent this graphically so what i can do here is just click our graph button here let's just look at the ratings of the movies versus the different ratings so i'm going to put ratings on the x-axis and on the y-axis i'm going to put the movie ratings here and we can see oh oops i'll just closed that let me do it again x-axis rating y-axis movie and we can see here tbma for the movies is the most common rating tv 14 second most common tv are third most common and tv pg fourth most common one thing i can do here just if i want to zoom in and look at a subset of the data get a better understanding of the compare and contrast here i can zoom in on a smaller set and then all i have to do is double tap and get back out to the larger data set so let's go back to the base data set and look at another relationship understand our data even more slice and dice it up a little bit more we see here we have the different countries these are the countries that the data sets are coming from in the bottom right corner i can see here i have 7787 rows in here and now if i apply a filter to country and let's say i just want to look at the ones that are from brazil so i'll do contains brazil we can see we filtered our data set down just to the brazil values or where where or values that contain brazil in it and below if i go here we see we've generated the equivalent code for that filter and we'll also see that there are 88 values left in the data set so we can see that we've shrunk the data set a good amount by filtering down to that filtering down to the brazil values i can remove that filter as well so let me remove that here and we're back to the base data set and let's look let's say i want to look at i want to see how many values how many movies or slash tv shows i have from each country all i have to do is pivot again actually i'll go back to this pivot table here and i can edit it so i'm going to change this here to to country i'm going to get rid of this and i'm going to change this to country and count and now here i can see for each country what is the amount of titles we have from there and if i want to make this a bit easier to view i'll hit change this to descending order and so now i can see united states has the most india the second most united kingdom the third most et cetera et cetera and below we have again the code for that pivot table and we have the code for that sword we did that sorting and descending order um so there's a lot of great analysis we can do in the tool mito is a really great way to slice and dice your data look at different patterns look at different behaviors in the data and understand how you want to proceed with the data thanks for checking out the video hope you really enjoyed it