ChatGPT Code Interpreter - Analysis Test!

I'm not going to review it I'm just going to hit run holy crap you just gotta analyze stuff open AI just Unleashed this new beta feature called code interpreter so if you're on chat GPT plus you go into your settings beta features code interpreter make sure this bad boy is turned on because you're going to want to check this thing out I couldn't resist I wanted to Dive Right In yesterday I was met with a bunch of uh too busy warnings whatever today I'm able to get it in rock and roll and I'm going to take this thing for a test drive so here's the deal I'm an Excel guy through and through I've used Excel to accelerate my career over the last 20 years and I'm currently finalizing the last module of my excel4 Academy which is Excel for business this is geared towards professionals who use Excel on a day-to-day basis and they want more out of their career and they want to use their Excel skills in order to help them get there one of the modules or one of the areas of focus in this training is going to dive deep into each department within a company starting with Finance going into HR operation sales marketing Etc and covering very typical scenarios that one might encounter out in the wild using Excel well I wanted to take one of my concepts for one of the tasks that I'm going to have people work through using power query to automate and ultimately productionize this exercise and I wanted to feed it to this new code interpreter and see how it could execute how well it does so let's do it let's Jump Right In I'm going to go ahead and pop down here and I'm going to show you what we've got this is just very high level stuff here what I have is completely fictitious data that I've made up this is an employee roster you've got a handful of employees first name last name Department very basic Excel file here I've got this June reimbursements data now we've got a month column you've got a name it's last comma first and you've got this reimbursement amount got some hanging blanks like this is if you were to like print out a statement or a report from some system where you've got all this stuff recorded the idea here is that basically these are employee reimbursements that are due for employees who are using a company card or their card or something like that right so nothing too crazy here and then lastly what I want to be able to do is compare this to a budget now this is just a Bare Bones budget assuming that this is just departmental reimbursement budget throughout the the year broken down by Department by month and these are all hard-coded values so again I went super dry on all the formatting I wanted to keep this very simple without being too obstructionist because I don't know what's going to happen I honestly don't I want to make sure that we're at least setting ourselves up for some level of success and if you you need to neuter your files a little bit before playing with this go have at it and that's what we're going to do so if I were to attack this in Excel and let's just say that the exercises hey we've got this spend by employee in this file and I want to see by Department how are we comparing to budget so I want to have a budget comparison by a department for the month of June to see who which department are they spending on above below where are they spending relative to the budget that's it that's the exercise so what you're actually going to have to do is combine these three files in order to execute that if you were to set this up in power query it would probably take a couple minutes to set up and then you should be able to execute it to push of a button we're not going to do that today we are going to Instead try to get chat GPT to do this for us so that's what I'm going to do I'm going to go ahead and jump right in to this this chat GPT again if you go to your settings you go to Beta features make sure that this guy is turned on and you should be able to see it I'm going to go to the gpt4 code interpreter beta all right now because you don't need to watch me type I'm going to go ahead and snap my fingers and now we've got our prompt filled in the only thing left is to attach our first file which is going to be our June reimbursements okay now I'm going to review the request for you just so that you can see how I prompted this I am not a prompt engineer I don't claim to be one I don't play one on TV but for you today I'm just going to kind of spell it out basically at a high level what it is I'm looking to do I'm attaching a file with employee reimbursements and this is file number one of three that I'm going to want to compare and analyze the second file is an employee roster which will provide the department each employee belongs to the third file is our budget for reimbursements by month by department for the year the output I'm looking for is going to be a table showing Department spend for June compared to budgeted spend for June with variances both in dollars and percentage please be sure to include a total row at the bottom so that we can see the overall impact please confirm that you understand the task and I will provide the additional two files so let's hit send fire in the hole and we are off to the races yes I understand the task we have three files the June reimbursements file which you've already uploaded this contains actual reimbursement amounts the roster and the budget we will use these to do these things awesome get ready to go alright great so now I'm going to load the employee roster right there and I'm just going to say perfect here is the roster no need for formalities thank you for providing the employee roster file now could you please upload the budget yeah it's asking for it it knows what's coming so here's the budget open and there you go enter and We're Off to the Races now I'm not going to do any editing at this point I want you to see in real time kind of like how this thing is operating and so far it's saying hey thanks for the files I'm going to go ahead and do all of these steps loading inspect join the data so you can see that summarize reimbursements data by Department extract the budgeted amounts compare the actual spend and provide the variances so let's start by loading and expecting the data from the three files so it's done that now it's going to say that it's successfully loaded but it looks like there's several rows of non-data so you saw the files right you saw that like I had it intentionally formatted a little bit wonky there's headers there's in some cases some erroneous like just added columns for no reason there's like hanging subtotals uh summer positioned at row four to start somewhere positioned at Row three so I intentionally kind of made the files a little bit different in each instance so we'll see how it kind of works through that so it's been loaded several rows of non data at the top blah blah blah for the reimbursements file we'll use it's moving fast so if you're interested in Reading every single one of these things by all means slow down the video pause the video check it out I'm not going to read the entire thing but it's saying now that the date is clean let's merge reimbursements the roster data it went through and loaded the new file it appears that the merge operation did not result in any matches between the employee name okay so what's going on here it could be due to the formatting differences or discrepancies in the name is written let's check the unique names so it's like it's identifying issues and it's pursuing Solutions so it's not just stopping and asking for confirmation or asking for me to change anything it's just muscling through the issue is that the order of the first and last names are in the play name or reverse between the two data sets that's correct again I intentionally did that one column had or one file has last name comma first name one file has first name and last name in separate columns so it's identified that and it's doing whatever it needs to do let's extract the budgeted amount from June so now we're in the budget file it seems there's an issue with the date format in the budget file the the date might not be recognized correctly due to its formats and again I formatted things however I kind of felt so the date formats are just Excel date formats which is a meme in and of itself but let's see what it's doing let's use the correct path format oh oh we're already producing results here look at this we've got a department Finance HR marketing operation and sales we've got amounts we've got totals we've got variances of recover so it's already producing this thing as you can see the actual system was higher in the budgeted amount with total variance okay sales HR marketing Finance performance spent more than budget while they operate look at that it provided a summary analysis this is freaking incredible now okay here's the deal uh so I'm gonna look at this and I go okay that's cool it it did it it did it very quickly and that's awesome um I'm always going to be skeptical first and I'm going to want to somewhat validate some of these things but at the same time I want to see okay all of that work that you just did right what I teach in the excel4 academy is going to be a lot of uh how to not only go through the steps and identify how to clean recognize what needs to be cleaned how to set it up but then to systematize it or productionize the automating of it using power query okay so all of that gets stored as a procedure that you can execute and it's really easy to refresh things so this thing I I'm gonna suspect that I should be able to say okay thank you for that can you go ahead and now convert this whole exercise into like a python script that I can execute on my own so that I can go and run it on my own machine a couple of caveats here I just want to make it very clear also this is completely fake data at this point in time in the way things are like I would never feed company data straight into this thing some people I mean it depends on the nature of the data but in general I would not be comfortable doing that at least not right now not in this instance I think that's probably the direction we're going is once you can start get companies comfortable doing this sort of thing then I mean the power just goes through the roof so let's go ahead and ask for our python file boom okay so it's going to go ahead and do that and let's see what it does you can install them using python python libraries okay here is the python script check it out it's coding and it's notated very nicely so it's defining the file paths load the data from the three files clean the reimbursements file clean the roster file reverse the order of the names in the roster file merge the data group the department calculate the total reimbursements extract budgeted amount for June merge actual budgeted data calculate variance in dollars calculate variance and percentages calculate totals to append the data frame oh my goodness then we're going to print the comparison total shall we shall we oh man I am anxious now let's see here I'm going to copy the code and we are going to grab our Jupiter notebook and I'm just gonna freaking plop it right in there it is I'm not gonna review it I'm just gonna hit run holy crap are you kidding me that was a fraction of a second mother of infant baby Jesus okay okay so I said I was skeptical um of course I still want to check this stuff out but I've got a Target to shoot for and this is amazing all right so let's go ahead and I'm gonna do this guy and I'm going to actually you know what I'm going to minimize this because I want to just start working in the Excel files real quick so we've got a reimbursement data and what I want to do in order to check this using Excel like super fast I'm not gonna like productionize this or anything like that I just want to do a real quick check on the totals so first things first I'm going to go to my employer roster I'm going to convert these names equals last name and quote comma space quote and first name this is going to create an index for me so that I can take this and compare it to my reimbursements and let's see here we're going to go to Department equals xlookup the name and I'm going to compare that here and retrieve the department Finance perfect that works now I can take this and go to my budget and we know we got June here I'm going to go ahead and say okay June actual sumifs one of my summing I am summing the reimbursement amount and I want it to be for the Departments and I'm going to find the department over here that's my total I'm going to sum this total now I want my variances dollar variance percent okay and this is going to be our budget for June was this minus actual 868 that number sounds familiar from the output and then I want this percentage to be from the budget okay copy down haste moment of truth let's bring back up our Jupiter notebook and here we go Finance oh I should probably make this so that I can see this alt wff and let's come over here so Finance actual okay this ties um the variance is 868. now that's interesting so it it calculated the variance differently than I would that's a note like that's something that I could probably Define in the term but it's showing like a variance of a positive amount meaning it's spent more but when you're dealing with expenses you want variances to be reflected as a negative if they overspend so that's just something I could Define I'm sure that it would be able to figure that out uh this one's 1962 that's there 655 there 1894. oh my goodness gracious 15 overall this thing it did it it just did all of that it did all of that and it did it very quickly that is super cool that is so cool I love where this is going um now I'm trying to think of like practical application and so the way that I'm going to set this up in my training program is to make sure that you've got uniformity and like how you're accessing and storing files making sure that everything all the parameters are set so that it can easily refresh when the data is updated you can absolutely do the same thing here because at the end of the day we're just dealing with code to systematize these things I think this is awesome I think you should check it out if you haven't already um I don't know what else to say I'm actually a little bit speechless but this is so cool what a time to be in this field do not be afraid of this stuff embrace it because you got you're gonna have to learn this this is going to make us more powerful but we also need to be responsible and understand that those that are going to wield the most I think influence in this situation are going to be those who understand where they're aiming and the basics of how to get there so that they can keep the machine in check that's what we need to focus on so hey I hope you got some value from this if you did consider giving me a subscribe or follow or whatever I'm active on LinkedIn Tick Tock and here so anyway that was fun hope you had fun too thanks for watching

Channel: Matt Brattin

Views: 1,026

Keywords: chatgpt code interpreter, chatgpt code interpreter plugin, chatgpt code interpreter demo, chatgpt code interpreter data analysis, chatgpt code interpreter data, chatgpt code interpreter examples, chatgpt code interpreter plugin demo, chat gpt code interpreter how to use, chat gpt code interpreter how to, chatgpt code interpreter csv, chatgpt code interpreter python, excel, excel analysis, chatgpt code interpreter excel, excel file code interpreter

Id: l135zCtPhOA

Length: 16min 14sec (974 seconds)

Published: Sat Jul 08 2023