You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Similar to #978, the convention requires the beam dimension for many variables, but for the echosounder models we support many of these variables do not change over ping_time during a deployment or survey, since they are properties of the transducers. This dimension was artificially created in the current code.
Since ping_time is a major dimension that data is being concatenated on, this repetition is costly when data is loaded into memory (though it compresses well). I think we should consider removing this redundant dimension.
@emiliom : tagging you so that it goes with work in #1056.
The text was updated successfully, but these errors were encountered:
Similar to #978, the convention requires the
beam
dimension for many variables, but for the echosounder models we support many of these variables do not change overping_time
during a deployment or survey, since they are properties of the transducers. This dimension was artificially created in the current code.Since
ping_time
is a major dimension that data is being concatenated on, this repetition is costly when data is loaded into memory (though it compresses well). I think we should consider removing this redundant dimension.@emiliom : tagging you so that it goes with work in #1056.
The text was updated successfully, but these errors were encountered: