Gregor Lenz

Sounds good 👍 thank you

should I merge?

oh I see, so for some datasets we still needs to decompress the data. I think that makes sense and a 5x speedup is definitely worth a little bit of...

thanks a lot @fabhertz95 !

@fabhertz95 this starts to look excellent 👌 the readme is great and I'm pleased to read that we can make use of lots of torchdata code. Can you explain to...

A performance hit of 11% is worth it. This is great stuff @fabhertz95 . Should I merge?

One more thing, is there a reason we're not using this anymore https://github.com/neuromorphs/tonic/blob/beef9808639693762893b9f17106a6d9ea61bc74/tonic/io.py#L157

Like in the old version

Sounds good 👍 merging now like there is no tomorrow

Hey @onlysubgroup, that's really interesting! So far I've only worked with neuromorphic sensor datasets, so I'm curious where this leads to. The fact that hdf5 files are available is already...