Hi all,
I suggest that we use a specific, schema-defined shape for the antenna gains, from "fastest dropped" to "latest dropped" axes (in killMS at least).
I also dropped the chunk values, as they seem like implementation-specific information to me? Potentially valuable but I would like to know what gain-specific operation it would serve. I believe that encoding the chunk start/stop values (t0,t1 etc) will be very valuable for combining heterogeneously-gridded gains, hence keeping them in.
Hi all,
I suggest that we use a specific, schema-defined shape for the antenna gains, from "fastest dropped" to "latest dropped" axes (in killMS at least).
I also dropped the chunk values, as they seem like implementation-specific information to me? Potentially valuable but I would like to know what gain-specific operation it would serve. I believe that encoding the chunk start/stop values (t0,t1 etc) will be very valuable for combining heterogeneously-gridded gains, hence keeping them in.