Automated Python Generation from the Mito Spreadsheet

Key Takeaways

hello all my name is krishnak and welcome to my youtube channel so guys today in this particular video we are going to see how to generate python automatically using the mito spreadsheet and here specifically we are going to use this library which is called as miter sheet this entire video has been recorded by jake who is from mito company itself and uh who will be entirely showing you with respect to all the ada part and how we can actually use this library in an amazing way so i hope you like this specific video and probably you watch this entirely it will be super important for you all in order to perform easy eda and to perform the other activities that we usually follow in the life cycle of a data science project so let's go ahead hey this is jake from mido i'm going to give you an update on some of mido's newest features and how you can use them to enhance your python data science and data analysis workflows so at a high level mito is a spreadsheet interface inside of a python environment right now we're inside of jupiter lab which is where mito works actually use mido inside jupyter lab which you can install very easily um the first thing we want to do is get mido into this spreadsheet and the important thing about mito is that every edit we make in the mito spreadsheet is going to generate the equivalent code down here so as you'll see if i filter if i do a pivot table anything from a spreadsheet it will generate the equivalent code in the code cell below so to get data into the tool i can do one of two things i can pass in a data frame as an argument like this or i can import from my files directly so i'm going to import uh this raman data here actually sorry i'm going to import some tesla stock data here we go and we see it imports my data from my spreadsheet into the data set into the miter sheet and it generates the code for doing that below so important before we move on is just to talk about how to install mido so these are the install instructions right here just these two commands run them in your terminal python dash m pip dash and pip installer might pip install mito installer and then python mito installer install so you run those and you'll get the miter sheet and then once you're inside of the notebook just import the mighty package and then maitoshi.chi calls the calls the interface here and all of this will be in links at the bottom of the video so now we have our data in uh one thing i might want to do is look at these different dates and let's look let's say i want to analyze this stock data by month so i can add a column here and i can use our month function and we're going to take a month out of this here and we get the month and see we're generating the code for that operation we just did so now that we've got the month value let's say i want to do a pivot table giving me the average open value per month so i'll click pivot we will do here actually before i do that i can go back and i can rename this column here as a month again we're generating code for these steps now i'm going to do a pivot table i shall just go here and as my row i'll put month and as the value i will put the open value and then we'll attach the mean here and we see this data is a little bit messy but we'll come back to that in a second and as we go down here we see we generated this code for a pivot table so again making the the pivot table in here generates the code here we see this is a good chunk of code that might take some time to write yourself but in mido you can do it in just a few seconds so i want to um i want to format this a bit differently to make it cleaner so let's just make it so it's to two decimal places so we see we can do that we can format our data frames like you would in a spreadsheet and the nice thing you can do with formatting or any of the data is you can export that data to an excel file so you can take your you can import files or you can import data frames and you can take the output of what you've done in mito and export it back to excel so a really great translation layer to go between python and excel is one of the great ways might it was used so now we're back at this pivot table we have these values let's say i want to do a graph of these values so i'm going to click the graph button and we're going to look at the months and then the mean open values here and here's a nice bar chart showing that value so we can change that to i think a line graph would do really well here there we go here we see a line graph showing the values we could also do a line graph just based on the base data set let's say we want to do a graph here and we just want to look at the date and then the high values there here's a nice chart showing that as a line graph there we go so multiple different line graphs we can make them on large data sets or small data sets and the nice thing we can do is we can copy this code as well so i'm going to copy the graph code here and i'm going to run this cell to edit the data frame with the edits we've made and then if i copy this code we get this plotly code so this is the graph code and we see this is a good chunk of code that might take some time to write but in mind you just make the graph in a really simple interface and then copy the code you have it and i will run that and we can actually get our code in the notebook itself some other examples of graphs just want to show you real quick we saw bar charts already we have scatter plots we have these density heat maps and we have these graphs here and i'm going to go back to this and so now that i go back to this data set um here let me just show you some other quick functionality tools so we've talked about adding columns and formulas we've talked about pivot tables and we've talked about graphing some other nice things you can do is we could do some filters for example so we have these values here let's say we want to say that just values greater than 20. we can do that you can see it edits the data frame and we know that there is a filter on because of that icon and again we get the code for that filter down here i'll close that we could also do undo and redo steps we can import and export files we can add columns we can delete columns so i go to this column right here delete now it's gone i can undo that step bring it back and these undo and redos will be recorded in the code below we can add console to the console pivot tables we can merge two data sets together we have different types of joins we can do lookups lefts rights inner outer unique and left unique and right we can deduplicate data sets to get rid of any duplicate values and we can graph and then we can format which we've already talked about so there's lots you can do in mido it's a spreadsheet inside python and again everything you do in the spreadsheet is going to generate the equivalent code in the uh in the cells below so i hope you check out mito and hear the install instructions one more time i'll link the documentation at the bottom um for you know the more the full install instructions and in our documentation you'll see information about how to use all the different features but yeah i hope you check it out and get some good use out of it thanks so much