Web Scraping with Python and ChatGPT

hello everyone it is Tuesday April 11 2023 I hope you're doing well tonight I'm using this camera let's move this over a little bit oh hey back to the old cam back to the olden days hey we're gonna do a quick stream tonight with some live coding try to figure some stuff up one of my web scraping let's figure that out right here right now um I'm gonna go ahead and share my screen don't have anything set up the way it should be but we're just gonna roll with it oh that's not right here we go there we are maybe this one's better okay so here's my chat here's my terminal let's load up into my repos twitch stream projects is it there oh wait repos there we go twitch stream projects let's make a new directory called band member uh graph I don't know and then we'll kind of activate kaggle 2023 and I'll open up Jupiter lab here and I want to show you guys what I'm thinking about doing tonight what's going on Josh says not much how are you doing tonight please give us a video on Transformers I've been thinking about it I got a understand them better in order to explain them better but maybe that that'll be motivation um so I was um listening to an interview recently with someone who was in a band and I happen to go on to the Wikipedia page and I'm just gonna go to a band Nirvana as an example and I've noticed this a lot um of musical groups that have band members on Wikipedia have this sort of same setup where it'll actually render this sort of an image of the band members and doing machine learning course would practice everything uh Albert thank you for the chatting any more NLP video coming up maybe any requests that you put here it'll sway me towards making a video on that topic so but right now we're looking at this stuff this is um just Wikipedia data right so this is these are the band members from this band and the years they were in the band but this rendered image down here actually uses a specific template it uses this template where you have this sidebar timeline title and then these years color and members now I need to I haven't delved into this much but I was looking at this band Nirvana's timeline to try to understand how it works and how we could potentially scrape it for a bunch of different bands and see what sort of overlap there is uh by the way what's going on with twitch is my twitch stream going yeah hey Rob what up Rob hello Rob yes some people are stuck on the Titanic data set let's go here to make sure I'm seeing my other my other window shows there's one viewer or maybe there is only one viewer okay it's cool that's fine um uh close all these windows okay so let's bring it back up let's bring it back up this is the source code for the Nirvana site right and I'm trying to figure out how we can scrape this for a bunch of different bands so maybe one of the first things we can do is just go to this Jupiter session create a new notebook in here um and see if we can scrape it so um if we go to here uh if we import requests scrape uh bands Wikipedia timeline member how did my change my Jupiter theme I installed uh the Dark theme and then I you go here into the drop down and you can change to different things like solarize dark and that's what I have installed I have a whole um I have a whole video on YouTube that you could check out on my Jupiter setup um I know there's a Wikipedia a Wikipedia um package which I might already have installed let's see if I have it installed let's import let me know if this is big enough for you guys to see import Wikipedia oh it's not installed so let's go to my content environment and pip install this blue screen is bad for the eyes is this blue is this bad for me all right let's see if this works I personally would use select to relax let's see what that looks like what is this theme oh selectolax is uh HTML5 parser with CSS selectors okay so should I try this out so this will parse the HTML let's look at the Wikipedia the Wikipedia um package first and just see what what comes up if I yeah this is an old one you can tell because it's using python 2 stuff um but just to just to see what I'll do if I do summary on Nirvana it'll just give me the summary just the top part if I do Dot Page then it gives me this page and then I have to do like dot title on this what is it dot content will give me everything and let's see if there's oh this is not the band though Nirvana and there we go what we want from here is the timeline uh so this has content I can set the length and then I can also make it to give me a summary but let's let's use selecto lacks like has been suggested so let's pip install this selectolex it's an alternative to beautiful soup okay well I like to learn new things so let's check it out from selecto last parser so let's see if there's a selectolax Wikipedia example so they have this HTML parser is that like the main part of it it looks like this is the main thing detecting coding use meta okay so from selecto clat LAX parser import HTML parser and if I get this URL right then I could at least use a Wikipedia to give me the URL so I still need to pull the URL because this is just gonna parse HTML so if I do requests dot get my URL then I have a response and my response is going to have what is it going to have let's check it out requests all right so we have a response status code which is 200 but we want to parse this okay so there's content in here all right this is the HTML so we're we're basically pulling this URL we're using Wikipedia to give us to search the URL for the band Nirvana we're getting the response using the request package and then we're going to use selectolax to actually give us the um to parse this HTML so HTML parser let's give it this HTML and what other parameters can we give it looks like that's mainly it um should we call this they like to call it the tree then from this tree how would we get to all right so this would just be the title the text in the title CSS first bad CSS selectors CSS all right so I search for test and nothing came back right uh hello do you mind if I ask you a question that I'm having some challenges with the Formula One data set from kaggle I don't really know that data set but you can ask a question um right once you get the request you get the text attribute for it and use that to create a tree you can isolate nodes by CSS class so I don't know that much about CSS how do I get to how do I get to then this timeline can I search her timeline node span I don't know what that means I'm just just searching blindly here any CSS matches let's go to the source of this page because I can see it here this is sort of what I want to get um so pull Wikipedia source in Python all right so this is just saying they use a Wikipedia package that's not what I want is there a way to get the source from from this source categories oh these are subcategories that are in it um content images original title page ID summary I want to start learning DS should I start with math or python a little bit of both sprinkle in one of each maybe sections section this is a method um so this would be members or I guess it's called timeline nope nothing of that title see here it has a tag it has a hashtag tag timeline and there's a standard for for Wikipedia timeline yeah this is what I was looking at before sidebar timeline basic usage for bands oops I don't want to edit this timeline wrapper what does that mean Vinnie basically I want my target as our array like racer this where racer is part of the features but I want to get what position qualified I got got into the race and how many points he won in that race sounds like you're gonna have to do some data wrangling do you remember recommend any math specifically I'd imagine statistics in linear algebra yeah definitely take an intro to statistics course or read a good book on it and um linear algebra you got the two main ones see this is looking for DOT post oh so I need to like make us I need to understand this see this is like just the raw data that I want to be able to parse the timeline here it gives the colors background colors line data so then it's showing where things started and changed I could basically make this into a data frame I could write the code to do that maybe that would be a better use of our time like how do we take this timeline uh try micro masters from edx offered by met yeah that sounds like a good one immediately I'm on mobile so you can't be 100 helpful oh it's it's cool it's cool so if this is our format that we have and this is just a string let's say we're able to extract this um it's basically like a dictionary except without [Music] um sort of sort of so we basically have to make a parser I think for this we have to parse each line it's not going to be an easy way to do this because then it gives each bar a name as start and end what is this a start and it end and then this is the color starting to end but this doesn't actually give us a start and end date so let's see how this actually looks when it's rendered all right because these start and end from the beginning to the very end so how did how does it know even to set that get what I'm saying maybe that's because it's set up here period yeah so they have a from and a till this is cool oh as they also tell us the date format because it could be different parsing Wikipedia [Music] timeline yeah it's just going to give us the Wikipedia page for parsing uh with python don't support we're not going to support it's saying to use beautiful soup we know that we could do this find all P get text this will just give us this text it's almost like I want to make these into little sub data frames what do you guys think uh isn't that timeline is an image The Source said it is a PNG file yeah I guess they take this and there's something that renders this like it would be cool if someone could someone knows how to figure that out like how how was this rendered from that but I've seen this on a lot of banned websites like name a band name any band let's look up like Wikipedia actually at the bottom here it has list of alternative rock artists so let's go into this list this list American hot dog is that actually a band all right Arctic Monkeys but I think it they mainly show it for bands where they have a lot of change in their lineups but look they even have it for the spin for Arctic Monkeys even though most of the members have been members throughout their whole duration that they've been but there's still this timeline now if we go to the edit View we look at this band's timeline we see it has this tag of timeline it gives us the formatting and all this stuff did you say said PNG somewhere because I don't see that okay so timeline so basically I want to figure out how to parse this and they have the main band timeline and then the touring timeline and that would be cool to see what artists because then you could form a actual graph to show all the relationships between bands if there are some some bands might just be out there on their own right turnover Fleetwood Mac that's a good one I could look that one up hey man I found your tutorials on spark and time series forecast um your recent PhD grad in data science most your work is an ml Healthcare we're dealing with clinical data looking forward to attending more of your streams for viewers I'm looking for data science Rules okay nice someone needs to hire hermit Monk thanks for writing the message though that's nice I'm glad you found the videos helpful I don't know how helpful I'm gonna be tonight but we're gonna try to look into this stuff see if we can find something helpful easy timeline syntax no no no no no no see all this data on Wikipedia is so valuable but how do we actually get it to be usable okay so let's just say let's just say I was able to parse this so we write a file and this is going to be called bar.txt and it will have all of this data and then we just use pandas to read that as if it was a CSV we need to import pandas and let's make this separator just any space I guess now we've kind of parsed this first and last name I wonder how it works if there's multiple last names like this would not work with that um we kind of just need to parse it correctly but we're just we're just brainstorming here so header is none I believe and now we have bars right and then we're gonna write let's write a file called plot data and that's going to have this why is it giving me a hard time oh I need a double Magic and then we're going to read this in with the same separator and the header is none um expected certain number of fields in line two and saw a different number than expected I think that's because of this line being missed but we could also do low memory equals faults I think we can make it not raise if there's an error parser error suppress parser errors pandas on bad lines skip or error bad lines equals faults I guess that's an older version on bad lines let's say skip in line two saw 21 so it's complaining about here this is going to skip most the lines that's no good I guess it didn't skip that many it's just having a real hard time parsing but we could see that this is the bar column and it should match up with this column in the bars data um so basically two three four five this is going to be the end and this seven is just the color and eight is just the width all right so this is our data subset we're going to name the columns is the bar start I don't think we want the fourth column I don't know if this is useful at all right so stop here Rammstein oh I could look that up from the HTML source that section is a div tag and inside an image tag to the image of the chart oh it'd be cool to download all the images too um so do you know the tag the hungman nice name can you tell me the um the div name of this so I could actually parse it from here that's your job right now hey guys what's going on just arrived uh we're trying to look at Wikipedia Pages for bands that have these type of image data uh or sorry this these images which are rendered from timeline data of who is in the band when they started when they ended and also what instruments they played and I want to figure out if I could somehow actually scrape this data so that we have some sort of database Master database of all of the different [Music] um members of different bands and when they were members and if there are some guys who are in like multiple or guys or girls who are in multiple different bands it would be cool to kind of figure that out um but the homework assignment I have for people watching is to figure out for me how to get the section which in within this tree um that would pull out for any band's website their timeline data that's what I want to pull out the timeline data and someone said that there's a tag in there for timeline I could see it here hashtag tag so what do I do do I just do this you guys tell me because this actually has this tree oh geez CSS should I just do this bad CSS selectors that returns nothing this returns nothing so we need to figure that out scrape the data you need to crawl the edit page of the band on the basic page you'll only be able to scrape the PNG okay so I have to actually go into like a in order to see this one I had to go into the edit page so you're saying if I actually change this URL to be the edit page now I'm pulling from the Arctic Monkeys edit page let's comment that out now the tree the thing is that tree has now this HTML has basically this field this whole box how do I get the data that's in here how do I get the data that's inside of there right if I just go to Raw HTML probably give me a lot of stuff yeah I'm guessing this has all the main stuff but not the actual stuff in that field right am I crazy here happy Tuesday shanty you can do regex Max see in here I don't think it has let's see if I can find Timeline what type is this so I need like a string representation of this all right so the text timeline is in this raw HTML we've gone that far decode all right so what is capsane here's the X path to the text WP text box whoa what does this do what does this do look for ID equals so do I do tree CSS CSS first detecting coding input encoding CSS matches nope node text area who is that something all right so this has given us the source editor we're getting there we're getting there what about XPath what does that mean what does that mean uh hey Tom hey Rob what time is it there it's uh hour after you you're in Colombia it's 9 10 p.m it is 10 13 P.M so maybe we're just off by three minutes hey Rob WD or WYD how you doing shush we're trying to figure out how to scrape Wikipedia and I need you guys to tell me how because I don't know how um I've gotten this far so we have an idea and we're just going all over the place with it once I can figure out how to get onto a oh we're we're going to like the edit site of a band's web page because if you're on the band's webpage you see this rendered right but if you go to the edit page and then you go under tags uh no timeline then we could get this text which we could then parse into some sort of like a uh data like a tabular format of data for that band but I need your help now inside this text you have a timeline what is this hermit monk gave me all this stuff too okay send a web response parsa HTML okay we've already done all this well not all of it but some find the text area extract the text did you get this room chat gbt did you get this from chat GPT you did get it from chat gbt I knew it but how do we know it's okay so let's just try it this way chat gbt's way we do have the URL already all right now we have the text area content oh we should we should just ask chat gbt to parse this whole thing let's do it let's do it let's use it if it's here how do we use this um how do I get this stupid sidebar to go away okay so new chat um let's see if chat gbt can help us parse this data um so we're almost there so once we have this um we need the timeline where's chat gbt where'd you go all right how can I parse text to only take out the part of the text hahaha from a Wikipedia source to only pull the timeline portion of the um page the data looks like this and I'm using python an error occurred all right we already did this timeline div.get text okay so we need we could use suit to find the thumb t right thumb t right what is this thumpty rate ah so much text um so here we did just text area is this timeline div gonna work no no LOL amateur chat GPT user yeah I'm not great at it in a way that makes me happy I'm glad I make you happy hola what's up we're doing a horrible job of trying to parse this data welcome and thanks for hanging out with us um let's do this let's ask it like this I have this code that will pull a wicked pedia source text and I want to parse parse out just the section containing the timeline I think I'm just a bad prompter I'm not good at this get the text from the text area content variable I can use regular Expressions oh okay let's see if this works so this regular expression should pull everything after timeline uh re is not imported and that didn't work maybe because it looks like this timeline with dashes let's say that didn't work that's great but the regex didn't pull it correctly the uh timeline section header looks like this and let me just paste all this in I wonder if the timeline is going to be a different format each time tag timeline um you can modify the regular expression pattern to look for the string three musicians timeline but I don't instead of timeline yeah I guess there it says timeline here has timeline with Dash is should I just look for anything with timeline without this without this no that doesn't work thing with that Wikipedia page editor is wikitex which is compiled into HML wrong header your browser doesn't have the dev tools like right click and inspect I can right click and inspect but then what do I do what do I do I'm inspecting this but then within this is just a bunch of like a blob of text I'm pretty sure person turquoise raising waving hello Mario you have to use a API to get the wiki text API parsing Wiki text what is this what is this I'm so confused foreign welcome to chat how's it going um so let's go back here I did inspect this but let's look at the timeline so here it should it does look like it has the triple quotes and then timeline why didn't this regex work for it shouldn't it search anywhere in the text area content and find the matching and then give us a result so it just gives me a oh because it's let's see what what chat GPT said after this it ends with the string bracket bracket so do we want to find anything what can you change the regex to find everything within uh a tag like this so it always I think looks like this tag slash timeline and then it ends with these two brackets if you select it inspect you can literally hover over the tags and they'll be selected and you'll know which to add to beautiful suit for but I think what we're getting at is that if we did it on the main page the source code HTML page will not have the text once you click change Wikipedia turns a Wiki text into a picture yeah that's right Russian soldier so I think Russian soldiers answered what rocket is saying which is if I go to the main page and I try to inspect that it's just an image that's been created all right we're trying our new search oh cool but the thing is ah it broke because it found these two brackets afterwards let's go to this so it found these two brackets and it pulled everything in between these brackets but I wanted to keep going until here um this works except for there is a double bracket like this within this tag and I want it to ignore if there has been a new sub tag starting with rack bracket so for example it should pull all of this the text and not stop after let's give all of this I want it to not stop after this for the text below does the timeline section end with the triple quotes um doesn't look like it like this is a triple quotes just to give this the title uh to account for nested tags you can modify the regex to use a negative look ahead assertation that excludes BRAC bracket if it follows another bracket bracket okay so let's let's see if this works still this is pretty impressive okay so we need to do place this this is our old match and should we do this flags.all what did it just do here it kind of works I don't know what it ended up doing it ended up pulling both of the different timelines within here this first timeline in the second one but it still stopped after here you guys see what happened timeline section only exists in the Wikipedia editor I mean once you leave the editor depends on the tag I'm inspecting random Wiki pages on Chrome and it can parse the tags with soup fairly common python scraping tutorial already on YouTube well I'm not trying so I'm trying to get this this data which doesn't exist on the Wikipedia page here's the Wikipedia page it actually doesn't have anything to inspect so if I inspect this it's just going to give me this PNG file which isn't very helpful for me because I'm not trying to deal with this PNG file I'm trying to get this raw data which someone in the chat told me I need to go to the edit page and then parse that so we figured out how to go into the edit page so like let's let's do the Nirvana URL and if we take this URL and we have to append it with actually I don't know if it would just work to do this so this is just the normal Nirvana URL and I think if we did and action edit no that doesn't work so it needs to be like this there's probably some way that we could automatically switch this but now we're going to pull the Nirvana Wikipedia source code not the actual page and then we see that does have timeline in there we're going to use this code to actually pull out the timeline data see it does happen to work here it does happen to work there because this doesn't have double brackets within it that was breaking it another example what is the thing I'm trying to do I don't know just I'm trying to have fun here no we're trying to pull this out and make it into something machine readable so now that I have this let's ask chat gbt to be able to parse this bar data and this plot data let's delete all these that we don't want um and let's just give it all of this and ask it to try to parse this can you write python code that will take this text and convert it into a python dictionary maybe that's too much to ask yes you can help with that no no no no no no no stop stop chat GPT you're just writing it out um no can you write the actual python code that will parse text that looks like this what the f what the crap is it doing why is it saying what is it is it just being crazy wait what is that please is that part bar plot data CSS the bar so you guys can go to it here let me just send it to you guys here's the link to the source page of Nirvana here's the actual page Wikipedia page and the subsection prompt it in the third person what do you mean it's it's just doing random stuff with Device IDs and descriptions and IP addresses that doesn't work at all I need it to be able to parse raw text like this and put it into a dictionary it's not going to know how to do it come on here's the here's the python code to do it it's not CSS it's a Wiki extensions that converts data into a PNG it's as confused as much as we are Rob what are you getting your responses just text it's not HTML parsable I know Vinnie catch up here just kidding just kidding all right what it what the freak is it doing here so it is telling me do have this text up here called timeline selection so it's going to strip and split in lines so then we have one so now this is going to make a all right so let's try to see what chat GPT is trying to do um it took this and it split it so each line now is its own element of this list the length of these these lines is 81. 81 lines of text and we're going to try to parse each one all right that's a good start then we're going to create an empty dictionary where we will put the result I mean that makes sense too let's put this up here once we understand it we put it up there now we're going to go through each lines oh this is smart and let's just print this key value so it's going to split the split on if it has a equal sign into sub lists I like it and then if it's colors then it's actually going to do all this crazy stuff and it breaks key value if key value is is colors then it does this else if key value zero then key value one so that broke when the key value probably when it's just nothing let's just do LF key value zero equals then it's going to continue let's do if the length of key value equals one then it'll continue all right so now it was able to parse some of this data so basically it only worked when when did it work it only worked when there were these equal signs it doesn't work here because the colors it starts to split lines so we could do something like we could look for when it has colors equals and then start to populate a dictionary for all these colors we can look for it when line line data equals is an element and then start to populate the lines the bar data yeah we could do this we could do this we don't even need chat GPT for this that's how add an alif that pukes whatever the variable is that it's Error airing um but yeah but then I would need like a try accept or something what do I click good morning four times GM good late evening to you hey chai time data science what's up how's it going I saw your stream recently I saw part of it it looked like you were having a lot of fun pack sir what are you scraping using scrapey uh wait what are you saying are you scraping using scraping no we're just scraping what do I click in the website Bonnie plays what are you trying to do um all right so let's make this simple well nothing here is simple but um so let's just forget about all this because we actually don't even care about the colors um so we're gonna say if line if it contains a shoot if it contains if this is equal in line so these are like all the start starting uh start start of each section I think then we'll start parsing stuff so if it has that then we'll do a split here and make this our section I don't know send text to said awk or rig prep or something you'll and be done with it okay tell me how to do that there are a few elements like colors that should be turned into a dictionary so I think chat gbt had the idea but I don't think I gave it enough of the information ah there's an error so basically we wanted to parse out the bar data and the plot data can you have it parse out the bar data and plot data from this text and put them in dictionaries all right so it's actually going to use a regular expression to pull out the bar text and the plot text from here okay that's probably a better way to do it okay let's try this out so we have our timeline selection here text timeline so our text is actually going to be yeah here and then it's also trying to find it in here oh cool so it wasn't able to find the bar data but it was able to parse this plot data if I do make this into a data frame oh wait this is going to be Orient index or I can just transpose this so now we have this is the band member name there's start date and their end date and then the color associated with it which really doesn't matter to us there's a copy code button for chat oh yeah I need to use that um but this worked for the bar data but it didn't pull out the plot data Why didn't it work for the plot data it did so well in the beginning no it pulled from the plot data but it didn't pull for the bar data oh this just looks for anything that starts with bar and then also has text like this bar text so why didn't it work so were there any bar matches no there was none maybe I just need this to be nope this is good but it didn't correctly pull the bar data for some reason uh congratulations for your work too bad there's no subtitles on the live stream I'm Brazilian oh sorry Felipe um maybe we'll get there soon where there can be live translation so wait so now is it doing it correctly it's actually pulling out the metadata too let's try this example um and let's just re write text with the timeline selection name here look behind requires fix with pattern I get this oh now it's writing out all the data it seems like a lot of work too much why is it writing all this text again I heard that newer versions of of chat GBC will actually have an interpreter okay so it just stopped here what what the crap and you Prov provide just the code python code to pull extract this data when you see a line that ends with Z equal you should oh okay yeah yeah that's a good that's a good point I should tell it to say this don't print the actual text string just show the code also when you see it a line that ends with equals you should create a separate function you should create a section function to Parks parse that subsection is insane Chachi BD can do all this but feels at the most ridiculous things I mean it's kind of failing at this it's not like it's doing I'm kind of turning my brain off and just seeing how well it does and it's entertaining but it hasn't solved this problem yet but yeah it does fail with a lot it also saved me a lot of time from having to write parsing code has it really saved time yet unless it works if you understand how it works the results are not that surprising yeah it's kind of like a really good Wikipedia answer if it works sorry not Wikipedia um you know stack overflow and there's no way this code's gonna just work there's no way so what is this trying to do parse timeline and it will take this raw text and it just returned nothing so it didn't find even this womp womp that's a lot of colors it is kind of cool here how it's actually choosing the function this is kind of a nice solution to what I was trying to think through a few minutes ago which was if it determines that the key is a certain thing it'll just change what the parsing function is not quite getting what it is doing here it's probably just a glob of code that's already out there that's how I'd write a parser for this it also simplifies the problem for chat gbt Rob is a data scientist how important do you think learning Ubuntu instead of Windows as I am a fresh fresher in college um I mean you could just just uh um be familiar with working in a Linux environment like in the command line you don't have to be on a Ubuntu machine you just need to um be able to like SSH into a Linux machine and be comfortable with moving files around and and that sort of stuff try it out it feels miserably uh what are you saying ask it to list the colors without letters it will fail but it can swap together code you can fix later does OS matter in data science no I mean you're mainly sshing into like ec2 instances or some sort of Linux machine environment anyways so the OS that you're running on your local environment is probably less important but I mean if you're running that on your local machine then you're just going to get used to it I don't know I think this parsing is just making my head spin a little bit um maybe that was enough fun for one night what do you guys think thanks everyone for hanging out tonight yeah I definitely feel you people in chat are saying to learn Linux that if your local files are stored in a Linux environment and then you're going to get comfortable moving those files around you're going to get comfortable with updating your OS by doing like apt update installing packages in certain ways yeah there's a lot of upside to doing it although I've heard that um the the Linux within Windows is supposed to be good like the subsystem but I've never used it so I can't speak to how well it works um I seem to only only get notified near the end of the stream make sure you hit all the bells and whistles on the on YouTube and on Twitch so here are a few links if you haven't checked them out already so here is my YouTube channel here that that second Link's a better link then we also have twitch where I stream from time to time sometimes more exciting than others and then um we also have a kaggle competition going on right now so this is free for anyone to join and we're giving away two gpus back there for the first place winner do you see that 4080 if you join the competition you have until May 1st and if you win you get this and just by joining you get a chance to win this bad boy 30 80. so um I just sent the link to the competition the data involved in the competition is this ah no hey there this data um you can see on the leaderboard there are 137 teams already competing and they're trying to predict my sleep patterns I've provided eight years of My Health Data um and you you're trying to predict when I sleep how much I sleep each night so it's a very fun competition you should definitely join you need to go watch uh GTC session in order to be eligible but other than that just read all the rules in the page and join you can still join now it's totally cool and you can still be eligible um what else is there all right so I have a Twitter that I sometimes use that I'll put here and where's my thing what the what the and that's it it's just a little issue with net related things I can see your IP network other than one on your machine what do you mean it can't okay I don't know what you're talking about all right so that's it for me tonight thanks everyone for hanging out with me uh we had like 80 people in YouTube and 40 50 some in twitch tonight so that's pretty fun I'll try to be better about having better stuff planned next time but maybe we can figure out how to parcel this data or we can just be in an endless loop of asking chat GPT to do it for us and then it never actually doing it correctly all right thanks everyone have a good night I'll see you next time bye-bye

Channel: Rob Mulla

Views: 10,266

Keywords:

Id: 9rSPLkBoRqg

Length: 82min 10sec (4930 seconds)

Published: Wed Apr 12 2023