LightFM kills Jupyter Kernel when fitting model

Nobody has claimed this yet.

So I've seen this issue repeated a couple times but never found a solution. Losses other than basic appear to immediately kill the Jupyter kernel, it doesn't even attempt to start fitting it just immediately implodes. Additionally for some reason I keep getting this warning: LightFM was compiled without OpenMP support. Only a single thread will be used.

I've uninstalled and reinstalled it at least 4 times at this point and attempted to work through it on my own. I ran the test code on Google Colab and it ran flawlessly so I know its an issue with my environment.

So my question is does anyone have an environment they'd be willing to share where everything runs as it's supposed to? My data is fairly large and I need to utilize the ram I have installed on my system but I've been beating my skull on my desk for like the last two days and I'm running out of hope.

Start by comparing the failing local Jupyter environment with the successful Google Colab run, focusing on the OpenMP warning and which non-basic loss triggers the kernel exit. Done requires a reproducible environment description and a confirmed cause or a focused report that maintainers can act on.

Written by the indexing model from the issue text.