So that was the missing part for me! I did it from scratch (with custom Python script with gdstk and shapely) (the GDS file does have the cells annotated, so not a big problem but still). I was thinking about scripting the "trace net" tool in klayout but decided that's going to probably bring its own can of worms...

Very cool seeing someone completely naive going into this :)

If you want to read more about a bit more... cheaty way to do this, I have written about using formal verification machinery to straight up force the solution out of the netlist here: https://atx.name/electronics/asic-re/ . Could be a bit of an infohazard, but I think journey is the goal and yours was certainly more educational :)

Ah yes, that matches my observations. It kinda sees that the stuff it is looking for is there, but does not see enough detail to actually notice that not only there is an endcap in the way, but the mug is also rotated the wrong way to sit in the holder.

It feels like the "r's in strawberry" effect where the models do not have enough introspection into the raw input data.

I maintain [1], which provides the models with the ability to render a screenshot from any angle and as far as I can tell, visually driven feedback does not work that well as this point. The models probably don't get enough of "lovecraftian garbled 3D model mess" in the training data or something...

[1] https://atx.github.io/OpenSCAD-Bench/

[1] https://www.glass-laser.cz/URANOVE-SKLO-c38_111_2.htm?page=2 [2] https://www.pacinekglass.com/produkt/breeze-uran