Closed wx4stg closed 9 months ago
adding the compression flag is a good thing, as it turns out :)
profiling commit d16e2e7, I get about a 1.5ms performance hit on an Apple M2 Max.
I don't think this is too terrible for the functionality it provides and it can be disabled.
Test data used attached.
for a rather busy one hour of data, it's a ~2ms hit if the data is already sorted, and a ~15ms hit if the data is randomly disorganized beforehand
Thanks, @wx4stg, for the changes in response to in-person conversation. Merging, and looking forward to community use of this.
I figured it would be a good idea to create an example script of how one could read in VHF source data, sort into flashes and grid using pyxlma, and then write out to a netcdf file for other tools to use.
...however, this currently produces netcdf files in the range of ~6GiB for 10 mins of data from TRACER. I think something I'm doing here is inefficient so I'll need to look into this more.