Couldn’t find this discussed from a search so asking.
If I upload a website to the Learn tab, will it scrape the meta data from images on the website?
Trying to make an image database and wondering if I can use each image’s meta data or if I just have to make a database of images and image links.
Thanks!
It usually does not. But you can see exactly what it scrapes by clicking into the webpage on the Knowledge Base. After the webpage is successfully uploaded, click on it in the Knowledge Base and it will open an interface that lets you see all the content the webscraper picked up.
For example, when I add a webpage from my blog about ai filmmaking I can click and see exactly what is appearing in each chunk. Links are appearing but the image alt descriptions are not.
