can you do a walkthrough to read a given url of news web page to come up with context of the news and come up with a summary through LLM, ideally using Ollama
Thank you for this video/post. I am trying to use a locally running llama model (lama3.3) running under Ollama. Running the python code complains that the model is not found, but continues to parse and summarize the articles. Can you suggest how to work with locally installed models like those that are Ollama based?
Hi, perhaps this article can be of help: www.scrapingbee.com/blog/scrapegraph-ai-tutorial-scrape-websites-easily-with-llama-ai/ Let me know whether you were able to solve your issue :)
can you do a walkthrough to read a given url of news web page to come up with context of the news and come up with a summary through LLM, ideally using Ollama
Hi, thanks for your comment. I have seen this error before, is it possible you're running Python 3.12? If so, try running it in a venv with 3.11. Good luck! If you need more help, you can reach me on my Discord (link in description).
whenever I install scrapegraphai I am hit by AttributeError: module 'pkgutil' has no attribute 'ImpImporter'. Did you mean: 'zipimporter'? [end of output]
It’s a great tool for people without Python / BeautifulSoup skills. For most cases which are a bit more complex, the conventional approach is probably the best ;)
Thanks for your comment. It’s impossible for me to make a video about every possible use case. Instead I try to make videos which can be easily applied to other scenarios. In this specific case, it’s just a matter of modifying the prompt and the url and you’re good to go 😉
I keep getting this error "PermissionError: [Errno 13] Permission denied: 'wired.xlsx'" and I've even typed the code shown in your video exactly and still I get that error, any help would be greatly appreciated
Hi Meta, it looks like you're trying to write a file to a directory where you don't have enough permissions. Assuming you're using Windows, can you try the following? Navigate to your Desktop, right click on any of your files there. Click properties. Go to the tab "general" and read location. This should be something like: "C:\Users\thoma\Desktop" (where thoma is your username). Copy this location to VS Code. Now add \wired.xlsx and change all slashes for double slashes (you should now have something like this: df.to_excel("C:\\Users\\thoma\\Desktop\\wired.xlsx",index=False) Run the code again. Let me know whether this works for you. If not, feel free to join my Discord if you need more help (link in video description).
@@TomsTechAcademy Hi, I managed to get it to produce two results and now I'm having issues again, It is now giving me this message "Traceback (most recent call last): File "c:\scrapperai\main.py", line 26, in df = pd.DataFrame(result['pages']) ~~~~~~^^^^^^^^^" any help would be greatly appreciated!
Hi, I just successfully installed the module. Please pip install scrapegraphai, and not scrapeRgraphai as mentioned in your other comment. This is the reason your import doesn't work
can you do a walkthrough to read a given url of news web page to come up with context of the news and come up with a summary through LLM, ideally using Ollama
Thank you for this video/post. I am trying to use a locally running llama model (lama3.3) running under Ollama. Running the python code complains that the model is not found, but continues to parse and summarize the articles. Can you suggest how to work with locally installed models like those that are Ollama based?
Hi, perhaps this article can be of help: www.scrapingbee.com/blog/scrapegraph-ai-tutorial-scrape-websites-easily-with-llama-ai/
Let me know whether you were able to solve your issue :)
It extracted the titles, not the articles themselves.
Thanks for writing this
can you do a walkthrough to read a given url of news web page to come up with context of the news and come up with a summary through LLM, ideally using Ollama
Thank you so much for showing the demo！ I still have a question：does scrapegraphai support full stack crawling？ or it only works for one page
Thanks for demo-ing the tool
Thank you for your comment, much apreciated!
Great video thx !
Thank you Bugra! 😀
this video tutorial was suprising beginner friendly
Thank you for your comment Amanda, and glad my video was useful for you :) I always try to make my videos as beginner friendly as possible
ModuleNotFoundError: No module named 'fp'
Can you scrape data from Google maps?
When i am trying this i get the error asyncio error cannot be called
Hi, thanks for your comment. I have seen this error before, is it possible you're running Python 3.12? If so, try running it in a venv with 3.11. Good luck! If you need more help, you can reach me on my Discord (link in description).
whenever I install scrapegraphai I am hit by AttributeError: module 'pkgutil' has no attribute 'ImpImporter'. Did you mean: 'zipimporter'?
[end of output]
Hi, I think I had this exact same error - are you running Python in a virtual environment? If you run Python 3.12 you can try switching to 3.11
I just give chat gpt the html data as example. Give it what I want scraped and it’ll generate the code for me
Thanks for Sharing your expertise, that’s another way of doing it 😀
I found ScrapeGraphAI pretty useless, it doesn't except for the simplistic examples.
It’s a great tool for people without Python / BeautifulSoup skills. For most cases which are a bit more complex, the conventional approach is probably the best ;)
No module named 'scrapergraphai'
Hi, the name of the module is scrapegraphai (not scrapeRgraphai)
i get api error
If you receive any error, feel free to join my discord (link in description) and post your question there 😀
Video could have been more helpful if it was about scraping helpful marketing information like emails and phone numbers
Thanks for your comment. It’s impossible for me to make a video about every possible use case. Instead I try to make videos which can be easily applied to other scenarios. In this specific case, it’s just a matter of modifying the prompt and the url and you’re good to go 😉
I keep getting this error "PermissionError: [Errno 13] Permission denied: 'wired.xlsx'" and I've even typed the code shown in your video exactly and still I get that error, any help would be greatly appreciated
Hi Meta, it looks like you're trying to write a file to a directory where you don't have enough permissions. Assuming you're using Windows, can you try the following?
Navigate to your Desktop, right click on any of your files there. Click properties. Go to the tab "general" and read location. This should be something like: "C:\Users\thoma\Desktop" (where thoma is your username). Copy this location to VS Code. Now add \wired.xlsx and change all slashes for double slashes (you should now have something like this:
df.to_excel("C:\\Users\\thoma\\Desktop\\wired.xlsx",index=False)
Run the code again. Let me know whether this works for you. If not, feel free to join my Discord if you need more help (link in video description).
@@TomsTechAcademy Hi, I managed to get it to produce two results and now I'm having issues again, It is now giving me this message "Traceback (most recent call last):
File "c:\scrapperai\main.py", line 26, in
df = pd.DataFrame(result['pages'])
~~~~~~^^^^^^^^^" any help would be greatly appreciated!
You should take this down.... Their module cannot be imported... As they now charge
Hi, I just successfully installed the module. Please pip install scrapegraphai, and not scrapeRgraphai as mentioned in your other comment. This is the reason your import doesn't work