How can I scrap for images from the unstructured web using just requests and beautifulsoup4? There's no existing dataset for the images I wanted to scrap that's why I'll do it myself.

> back to /g/
Nobody answered my question there. I'm also curious how can I filter wrong images? Do I train a small subset of correct datasets with some ML model? I'm still ignorant about this stuff, that's why I asked here.