Amon Lahr

@AlexProgrammerDE Thanks for your quick response! After setting the `BALENA_HOST_CONFIG_dtoverlay` parameter to `wm8960-soundcard`, I got the following error messages: [dtoverlay_wm8960-soundcard.txt](https://github.com/balenalabs/balena-sound/files/5033548/dtoverlay_wm8960-soundcard.txt) Also I tried to connect via bluetooth and Spotify Connect...

Three days of not doing what I was intended to later I finally made it work! Not the most elegant solution but it works for my Pi Zero W with...

@tmigone You are right about the initial Spotify issues which turned out not to be related to the soundcard. Eventually I got everything to work using my additional container that...

To me this looks like the config was correct but there are still driver issues. Which hardware are you using? Tested the repo on my Raspberry Pi Zero W with...

Thanks for the tip! Exactly, basically I would like to chain the smoothing solves for different vector fields. I will take a look at your suggested approach and let you...

Yes, thanks! I will give it a try!

Hi, small update: I have managed to stitch together solutions with different vector fields; however, how to initialize the solver at the beginning of the non-first interval is still unresolved...

Hi @pnkraemer, thanks for following up! I did not continue working on this. As far as I remember, the main issue seems to be rather fundamental: 1. If one would...

It could be that with @naefjo and @alexhansson we encountered a similar issue. Maybe this also resolves your issue? https://github.com/cornellius-gp/gpytorch/commit/c7906a6fba16833367e145cd81fe6107e33d802a You can check out [this fork](https://github.com/cornellius-gp/gpytorch/compare/main...naefjo:gpytorch:feature/fix-memory-leak) for the details. We...

At the moment I am not quite sure about the difference between `LazyEvaluatedKernelTensor` and `LinearOperator` objects. Could you give me a pointer where to find under which conditions a `LinearOperator`,...