Nils Lehmann

Thank you very much for your reply @zou3519. I'm sorry I was not specific enough in my description. For the Jacobian computation, I am actually interested at an even more...

Yes, so an example would be a neural net that computes features at different resolutions or scales that subsequently get processed together for a final output.

That looks like a potential solution at some point, once convolutions are supported. I will keep an eye out.

1. If you have a preference for a standardized way of doing this, I can change all the datasets accordingly. 2. Given a standardized way of choosing bands, I can...

After spending some time on this I have a more general question about these types of datasets. And since to my knowledge there is not yet a `GeoDataset` that includes...

Yes I think #511 can be closed. But this PR still needs some work, since they are all time-series and additionally three of the datasets are veeery label sparse so...

I would be really interested in taking on this task!

Thanks for your fast reply. > * [USGS Earth Explorer](https://earthexplorer.usgs.gov/) - Many different datasets I first tried USGS Earth Explorer, but the GeoTiffs one can download there have 20x20m resolution...

> Why do you need to have the S2 tiles split up? (I think this will make the process much more complicated). I don't think I need the S2 tiles...

I would advocate for the first solution, since I have run into that issue as well, where a remote machine with GPU has not enough memory for all 140GB but...