xarray Datasets from CASA Tables

Constructs xarray Datasets from CASA Tables via python-casacore. The Variables contained in the Dataset are dask arrays backed by deferred calls to pyrap.tables.table.getcol.

Supports writing Variables back to the respective column in the Table.

The intention behind this package is to support the Measurement Set as a data source and sink for the purposes of writing parallel, distributed Radio Astronomy algorithms.

Installation

To install with xarray support:

$ pip install dask-ms[xarray]

Without xarray similar, but reduced Dataset functionality is replicated in dask-ms itself. Expert users may wish to use this option to reduce python package dependencies.

$ pip install dask-ms

Example Usage

  import dask.array as da
  from daskms import xds_from_table, xds_to_table

  # Create xarray datasets from Measurement Set "WSRT.MS"
  ds = xds_from_table("WSRT.MS")
  # Set the flag Variable on first Dataset to it's inverse
  ds[0]['flag'] = (ds[0].flag.dims, da.logical_not(ds[0].flag))
  # Write the flag column back to the Measurement Set
  xds_to_table(ds, "WSRT.MS", "FLAG").compute()

  print ds

[<xarray.Dataset>
 Dimensions:         (chan: 64, corr: 4, row: 6552, uvw: 3)
 Coordinates:
     ROWID           (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
 Dimensions without coordinates: chan, corr, row, uvw
 Data variables:
     IMAGING_WEIGHT  (row, chan) float32 dask.array<shape=(6552, 64), chunksize=(6552, 64)>
     ANTENNA1        (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
     STATE_ID        (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
     EXPOSURE        (row) float64 dask.array<shape=(6552,), chunksize=(6552,)>
     MODEL_DATA      (row, chan, corr) complex64 dask.array<shape=(6552, 64, 4), chunksize=(6552, 64, 4)>
     FLAG_ROW        (row) bool dask.array<shape=(6552,), chunksize=(6552,)>
     CORRECTED_DATA  (row, chan, corr) complex64 dask.array<shape=(6552, 64, 4), chunksize=(6552, 64, 4)>
     PROCESSOR_ID    (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
     WEIGHT          (row, corr) float32 dask.array<shape=(6552, 4), chunksize=(6552, 4)>
     FLAG            (row, chan, corr) bool dask.array<shape=(6552, 64, 4), chunksize=(6552, 64, 4)>
     TIME            (row) float64 dask.array<shape=(6552,), chunksize=(6552,)>
     SIGMA           (row, corr) float32 dask.array<shape=(6552, 4), chunksize=(6552, 4)>
     SCAN_NUMBER     (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
     INTERVAL        (row) float64 dask.array<shape=(6552,), chunksize=(6552,)>
     OBSERVATION_ID  (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
     TIME_CENTROID   (row) float64 dask.array<shape=(6552,), chunksize=(6552,)>
     ARRAY_ID        (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
     ANTENNA2        (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
     DATA            (row, chan, corr) complex64 dask.array<shape=(6552, 64, 4), chunksize=(6552, 64, 4)>
     FEED1           (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
     FEED2           (row) int32 dask.array<shape=(6552,), chunksize=(6552,)>
     UVW             (row, uvw) float64 dask.array<shape=(6552, 3), chunksize=(6552, 3)>
 Attributes:
     FIELD_ID:      0
     DATA_DESC_ID:  0]

Documentation

https://dask-ms.readthedocs.io.

Limitations

1. Many Measurement Sets columns are defined as variably shaped, but the actual data is fixed. dask-ms will infer the shape of the data from the first row and must be consistent with that of other rows. For example, this may be issue where multiple Spectral Windows are present in the Measurement Set with differing channels per SPW.

   dask-ms works around this by partitioning the Measurement Set into multiple datasets. The first row's shape is used to infer the shape of the partition. Thus, in the case of multiple Spectral Window's, we can partition the Measurement Set by DATA_DESC_ID to create a dataset for each Spectral Window.