scruffynerf

> @scruffynerf These are the pads I'm talking about. >  I wonder if one is a potential White, as in RGBW. But honestly these have to be some...

I just wrote the basics of a deepdanbooru tagger (using existing tf keras model) which does pictures. The idea of video analysis is interesting. But with the price of GPUs...

I believe the file refactor will allow things like this. Right now, it wouldn't be. Wait till the refactor goes live, and discuss more then.

``` selector: (//div[@id='tagline']/text() | //div[@id='description']/p[1]/text()) concat: "\n" ``` yields this output: ``` Her friend thinks everything is fine. Sam is the best friend of Jolie (from ). She gives her...

> an option to generate default thumbnails for selected videos (not on detail page so I had to click through them one by one to generate default thumbnails). I wrote...

Yeah, we began discussing this in Discord, and I agree with removing "Tag" names entirely, especially Autotag, which needs to be hidden from users who don't understand what it does...

While I agree in principle... The real world (that is discussion with multiple people on Discord... Many multiple people...) The tag icon is meaningless to them. They don't see it...

If you review the thread on discord, you weren't the only voice saying the above. It wasn't personal. It was factual and reflected multiple views expressed, not just yours. You...

> Any suggestions how writing a script like that would look? For someone who does essentially no scripting other than batch commands to copy/paste stuff. I'm going to be posting...

Ok, while I did a simple scraper that recognized mobile, then I decided tagging VR by looking at dimensions was worthwhile... and then I expanded to all aspect ratios... and...