Web Scraping With ChatGPT Code Interpreter: Mind-Blowing !!

[@TheAIAndy]

Your Most Powerful, All-In-One AI Writing Copilot: hix.ai/?utm=aiandy Thanks to Hix AI For sponsoring this video. Our sponsors help us grow the channel and keep making better videos for you. Thanks for your support!

[@PubgSpeed-wl8yo]

How can I get the full article? There is a lot more data inside the article, why do I need this superficial information? Do you know anyone who parses like this? Why not show how everything you need comes from inside?

[@VeVeVerse]

It's cool but I want to automate the process to where I can set everything up once, and it auto scrapes new information as it comes in.

[@thechaparral]

Perfect! Did you already get to do it on chatgpt?

[@danityvanityinsanity]

Can you create an ai agent to automate it or custom gpt?

[@ILONA-DAVID-lifelong-learning]

Hi, I have a plus subscribtion , but the Advanced Data Analysis is not visible for me (before I could see the Code interpreter, now nope)

[@DanielTheCunningham]

QQ: In the second example, you glossed over how you corrected the error. Can you elaborate on how you modified your prompt to solve the issue?

[@ClimbHighWithAI]

I need to do it with more than 5000+ products and also need description and price and etc how can I do it

[@Greencollarguy]

Total rookie question: what is the purpose of you doing this? Like the blog page what purpose does that serve? Building an LLM?

[@npane171]

This doesn't work with webpages that have information that is dynamically loaded, like zoominfo.

[@rocketgrowthstrategiesdigi4535]

On the tech crunch you could do a rewrite

[@PatrickCordero-j5m]

man your a savior

[@EricaTroni]

Thank you for this video! Is there a way to scrape product images as well as all info you’ve mentioned?

[@torarinvik4920]

Kudos to you for diving into coding! Scraping is imo a huge pain in the ass, so it's good to have tools to simplify it!

[@PubgSpeed-wl8yo]

Do you understand how to get a full article from parsing?

[@vladimirxack]

​@@PubgSpeed-wl8yo The same way bro used inspect element, you can do that too.

[@CyberMacs]

Great idea, thank you! :)

[@training7574]

Very useful and simple, thanks.

[@olieducationalvideos6203]

I think Bardeen ai can do this all with just a prompt

[@IrlymMylros]

It will amazing if you can show how to generate complete scripts for Cloud based operstions like complete company infrastructure setup in Microsoft Azure, scripts like Powershell or CLI. Or Microsoft 365 setup. Many thanks for your great videos.

[@gyangting]

Is there a way to webscrape without downloading an html file, i.e. just providing a page link

[@TheAIAndy]

not yet

[@william254]

There is. Use a python web scraper.

[@JeremyJullien-ry1pz]

This was really helpfull ! thx

[@ajithbaburaj2547]

Awesome🎉... Shorter video👌🏼

[@RenatinhoLicioniFilho]

I'm trying to setup a web scraper for facebook marketplace, can you help me with that?

[@japhethmutuku8508]

I can help you get with that. Let me know if you’d like more details or if you have any questions!

[@william254]

This isn't web scraping. You need to be able to do this automatically. For example, what happens when the data on the websites change? But it's a good alternative if you just want data from a few pages.

[@MisterSkraetsch]

Thanks Andy. Keep up the great work. :) I would be interessed in web scrapping for sites where you first need to LOGIN :) Cheers

[@mcombatti]

Even faster, you could just copy paste the site data using select all & copy, then ask GPT to create a CSV for the items found in the unorganized data(not page source) 🤫 take me all but 10 seconds to do the same as code interpreter can do within minutes, with less effort. Code interpreter is great, it's like a really big hammer... but thumbtacks don't require a hammer 😉

[@SamuelNickolas-yf7wh]

@mcombatti If it is not the page source then what did you copy?

[@kenzabel5394]

yeah but copy and paste inside chatgpt will make the prompt so long it will make it hard to read what you wrote in chatgpt

[@mcombatti]

@kenzabel5394 the context length of GPT-4o is 120,000 tokens - or about 300 pages of text. I haven't hit the ceiling yet. 🙏

[@mcombatti]

@@SamuelNickolas-yf7wh open the page and hit copy and paste... page source would just be silly. 🤭

[@mcombatti]

@@kenzabel5394 I typically am more concerned with the response as I know what the input was 🙏

[@beautytips911]

Awsome examples, can do one with Google my businesses

[@celestialresearch]

Pretty cool!

[@AdamBechtol]

Neat, thanks.

[@oliverli9630]

that's an interesting method. a normie can do that at home if they only want the first few pages. if they need multi pages, python coding still is a must

[@aarong9327]

So you're saying I'd need python if I wanted more in depth data on each of those headphones?

[@oliverli9630]

@@aarong9327 yup. programming skill still gives an edge

[@PubgSpeed-wl8yo]

@@oliverli9630 the conclusion is why this video was created, which does not carry complete instructions, which 99% of people need, why this video is needed, what is useful in it. You surprise me in the West, you don’t criticize anyone who sees the problem, you keep saying, bro, you’re handsome, this is useful. But in fact, this is of zero benefit, because it did not show how to take full content from inside the article. What he showed was just for the video and not for training.

[@JJBoi8708]

Do a video on open interpreter!!!

[@erikmeier3681]

why you are so fast?

[@Web.Scraping]

This is ok for small amounts of data, but for sites that have hundreds of pages of pagination this way is too laborious. In any case, this video will be useful to someone 👍

[@tanvirahmed1959]

awsome really