org.das2.qds.util.LSpec
Form a rank 2 dataset with L and Time for tags by identifying monotonic sweeps
in two rank 1 datasets, and interpolating along the sweep.
USER_PROP_SWEEPS
identifies the sweep of each record
DIR_INWARD
just the inward sweeps
DIR_OUTWARD
just the outward sweeps
DIR_BOTH
include both sweeps
identifySweeps
identifySweeps( QDataSet lds, int dir ) → QDataSet
identify monotonically increasing or decreasing segments of the dataset.
Parameters
lds - the dataset which sweeps back and forth, such as LShell or MagLat (or a sine wave for testing).
dir - 0=both 1=outward 2= inward
Returns:
rank 2 data set of sweep indeces, dim0 is sweep number, dim1 is two-element [ start, end(inclusive) ].
[search for examples]
[view on GitHub]
[view on old javadoc]
[view source]
rebin
rebin( QDataSet lds, QDataSet zds, QDataSet lgrid ) → QDataSet
rebin the datasets to rank 2 dataset ( time, LShell ), by interpolating along sweeps. This
dataset has the property "sweeps", which is a dataset that indexes the input datasets.
Parameters
lds - rank 1 dataset of length N
zds - rank 1 dataset of length N, indexed along with {@code lds}
lgrid - rank 1 dataset indicating the dim 1 tags for the result dataset.
Returns:
a rank 2 dataset, with one column per sweep, interpolated to {@code lgrid}
[search for examples]
[view on GitHub]
[view on old javadoc]
[view source]
rebin
rebin( QDataSet tlz, QDataSet lgrid, int dir ) → QDataSet
alternate, convenient interface which where tlz is a bundle of buckshot data (T,L,Z)
Parameters
tlz - x,y,z (T,L,Z) bundle of Z values collected along inward and outward sweeps of L in time.
lgrid - desired uniform grid of L values.
dir - =1 increasing (outward) only, =-1 decreasing (inward) only, 0 both.
Returns:
a rank 2 dataset, with one column per sweep, interpolated to {@code lgrid}
[search for examples]
[view on GitHub]
[view on old javadoc]
[view source]
rebin
rebin( QDataSet tt, QDataSet lds, QDataSet zds, QDataSet tspace, QDataSet lgrid, int dir ) → QDataSet
alternate algorithm following Brian Larson's algorithm that rebin the datasets to rank 2 dataset ( time, LShell ), by
interpolating along sweeps. This dataset has the property "sweeps", which is a dataset that indexes the input datasets.
Parameters
tt - The Time values corresponding to x axis position. If null, then use lds.property(QDataSet.DEPEND_0).
lds - The L values corresponding to y axis position, which should be a function of time.
zds - the Z values corresponding to the parameter we wish to organize
tspace - rank 0 cadence, such as dataset('9 hr')
lgrid - rank 1 data is the grid points, such as linspace( 2.,8.,30 )
dir - =1 increasing (outward) only, =-1 decreasing (inward) only, 0 both
Returns:
a rank 2 dataset, with one column per sweep, interpolated to {@code lgrid}
[search for examples]
[view on GitHub]
[view on old javadoc]
[view source]
rebin
rebin( QDataSet lds, QDataSet zds, QDataSet lgrid, int dir ) → QDataSet
rebin the datasets to rank 2 dataset ( time, LShell ), by interpolating along sweeps. This
dataset has the property "sweeps", which is a dataset that indexes the input datasets.
Parameters
lds - rank 1 dataset of length N
zds - rank 1 dataset of length N, indexed along with {@code lds}
lgrid - rank 1 dataset indicating the dim 1 tags for the result dataset.
dir - =1 increasing (outward) only, =-1 decreasing (inward) only, 0 both
Returns:
a rank 2 dataset, with one column per sweep, interpolated to {@code lgrid}
[search for examples]
[view on GitHub]
[view on old javadoc]
[view source]