def clean(**kw):
'''
Single-scale clean.
If the optional weight-table argument points to a valid weight table
(created by the psf worker) the algorithm will approximate gradients using
the diagonal Mueller weights assumption (exact for Stokes I imaging) i.e.
IR = ID - R.H W R x
otherwise it is a pure image space algorithm i.e.
IR = ID - PSF.convolve(x)
The latter is exact in the absence of wide-field effects and is usually
much faster.
If a host address is provided the computation can be distributed
over imaging band and row. When using a distributed scheduler both
mem-limit and nthreads is per node and have to be specified.
When using a local cluster, mem-limit and nthreads refer to the global
memory and threads available, respectively. By default the gridder will
use all available resources.
Disclaimer - Memory budgeting is still very crude!
On a local cluster, the default is to use:
nworkers = nband
nthreads-per-worker = 1
They have to be specified in ~.config/dask/jobqueue.yaml in the
distributed case.
if LocalCluster:
nvthreads = nthreads//(nworkers*nthreads_per_worker)
else:
nvthreads = nthreads//nthreads-per-worker
'''
args = OmegaConf.create(kw)
pyscilog.log_to_file(args.output_filename + '.log')
if args.nworkers is None:
args.nworkers = args.nband
OmegaConf.set_struct(args, True)
with ExitStack() as stack:
# numpy imports have to happen after this step
from pfb import set_client
set_client(args, stack, log)
# TODO - prettier config printing
print('Input Options:', file=log)
for key in args.keys():
print(' %25s = %s' % (key, args[key]), file=log)
return _clean(**args)
def jones2col(**kw):
'''
Write product of diagonal Jones matrices to 'Mueller' column
'''
args = OmegaConf.create(kw)
pyscilog.log_to_file(args.output_filename + '.log')
from glob import glob
ms = glob(args.ms)
try:
assert len(ms) == 1
args.ms = ms
except:
raise ValueError(f"There must be exactly one MS at {args.ms}")
OmegaConf.set_struct(args, True)
with ExitStack() as stack:
from pfb import set_client
args = set_client(args, stack, log)
# TODO - prettier config printing
print('Input Options:', file=log)
for key in args.keys():
print(' %25s = %s' % (key, args[key]), file=log)
return _jones2col(**args)
def dirty(**kw):
'''
Create a dirty image from a list of measurement sets.
The dirty image cube is not normalised by wsum as this destroyes
information. The MFS image is written out in units of Jy/beam.
The normalisation factors can be obtained by making a psf image
using the psf worker (see pfbworkers psf --help).
If a host address is provided the computation can be distributed
over imaging band and row. When using a distributed scheduler both
mem-limit and nthreads is per node and have to be specified.
When using a local cluster, mem-limit and nthreads refer to the global
memory and threads available, respectively. By default the gridder will
use all available resources.
Disclaimer - Memory budgeting is still very crude!
On a local cluster, the default is to use:
nworkers = nband
nthreads-per-worker = 1
They have to be specified in ~.config/dask/jobqueue.yaml in the
distributed case.
if LocalCluster:
ngridder-threads = nthreads//(nworkers*nthreads_per_worker)
else:
ngridder-threads = nthreads//nthreads-per-worker
'''
args = OmegaConf.create(kw)
pyscilog.log_to_file(args.output_filename + '.log')
from glob import glob
ms = glob(args.ms)
try:
assert len(ms) > 0
args.ms = ms
except:
raise ValueError(f"No MS at {args.ms}")
if args.nworkers is None:
args.nworkers = args.nband
OmegaConf.set_struct(args, True)
with ExitStack() as stack:
from pfb import set_client
args = set_client(args, stack, log)
# TODO - prettier config printing
print('Input Options:', file=log)
for key in args.keys():
print(' %25s = %s' % (key, args[key]), file=log)
return _dirty(**args)
def forward(**kw):
'''
Extract flux at model locations.
Will write out the result of solving
x = (R.H W R + sigmainv**2 I)^{-1} ID
assuming that R.H W R can be approximated as a convolution with the PSF.
If a host address is provided the computation can be distributed
over imaging band and row. When using a distributed scheduler both
mem-limit and nthreads is per node and have to be specified.
When using a local cluster, mem-limit and nthreads refer to the global
memory and threads available, respectively. By default the gridder will
use all available resources.
Disclaimer - Memory budgeting is still very crude!
On a local cluster, the default is to use:
nworkers = nband
nthreads-per-worker = 1
They have to be specified in ~.config/dask/jobqueue.yaml in the
distributed case.
if LocalCluster:
ngridder-threads = nthreads//(nworkers*nthreads_per_worker)
else:
ngridder-threads = nthreads//nthreads-per-worker
'''
args = OmegaConf.create(kw)
pyscilog.log_to_file(args.output_filename + '.log')
if args.nworkers is None:
args.nworkers = args.nband
OmegaConf.set_struct(args, True)
with ExitStack() as stack:
from pfb import set_client
args = set_client(args, stack, log)
# TODO - prettier config printing
print('Input Options:', file=log)
for key in args.keys():
print(' %25s = %s' % (key, args[key]), file=log)
return _forward(**args)
def _predict(ms, stack, **kw):
args = OmegaConf.create(kw)
OmegaConf.set_struct(args, True)
pyscilog.log_to_file(args.output_filename + '.log')
pyscilog.enable_memory_logging(level=3)
# number of threads per worker
if args.nthreads is None:
if args.host_address is not None:
raise ValueError(
"You have to specify nthreads when using a distributed scheduler"
)
import multiprocessing
nthreads = multiprocessing.cpu_count()
args.nthreads = nthreads
else:
nthreads = args.nthreads
if args.mem_limit is None:
if args.host_address is not None:
raise ValueError(
"You have to specify mem-limit when using a distributed scheduler"
)
import psutil
mem_limit = int(psutil.virtual_memory()[0] /
1e9) # 100% of memory by default
args.mem_limit = mem_limit
else:
mem_limit = args.mem_limit
nband = args.nband
if args.nworkers is None:
nworkers = nband
args.nworkers = nworkers
else:
nworkers = args.nworkers
if args.nthreads_per_worker is None:
nthreads_per_worker = 1
args.nthreads_per_worker = nthreads_per_worker
else:
nthreads_per_worker = args.nthreads_per_worker
# the number of chunks being read in simultaneously is equal to
# the number of dask threads
nthreads_dask = nworkers * nthreads_per_worker
if args.ngridder_threads is None:
if args.host_address is not None:
ngridder_threads = nthreads // nthreads_per_worker
else:
ngridder_threads = nthreads // nthreads_dask
args.ngridder_threads = ngridder_threads
else:
ngridder_threads = args.ngridder_threads
ms = list(ms)
print('Input Options:', file=log)
for key in kw.keys():
print(' %25s = %s' % (key, args[key]), file=log)
# numpy imports have to happen after this step
from pfb import set_client
set_client(nthreads, mem_limit, nworkers, nthreads_per_worker,
args.host_address, stack, log)
import numpy as np
from pfb.utils.misc import chan_to_band_mapping
import dask
from dask.distributed import performance_report
from dask.graph_manipulation import clone
from daskms import xds_from_storage_ms as xds_from_ms
from daskms import xds_from_storage_table as xds_from_table
from daskms.utils import dataset_type
mstype = dataset_type(ms[0])
if mstype == 'casa':
from daskms import xds_to_table
elif mstype == 'zarr':
from daskms.experimental.zarr import xds_to_zarr as xds_to_table
import dask.array as da
from africanus.constants import c as lightspeed
from africanus.gridding.wgridder.dask import model as im2vis
from pfb.utils.fits import load_fits
from pfb.utils.misc import restore_corrs, plan_row_chunk
from astropy.io import fits
# always returns 4D
# gridder expects freq axis
model = np.atleast_3d(load_fits(args.model).squeeze())
nband, nx, ny = model.shape
hdr = fits.getheader(args.model)
cell_d = np.abs(hdr['CDELT1'])
cell_rad = np.deg2rad(cell_d)
# chan <-> band mapping
freqs, freq_bin_idx, freq_bin_counts, freq_out, band_mapping, chan_chunks = chan_to_band_mapping(
ms, nband=nband)
# degridder memory budget
max_chan_chunk = 0
for ims in ms:
for spw in freqs[ims]:
counts = freq_bin_counts[ims][spw].compute()
max_chan_chunk = np.maximum(max_chan_chunk, counts.max())
# assumes number of correlations are the same across MS/SPW
xds = xds_from_ms(ms[0])
ncorr = xds[0].dims['corr']
nrow = xds[0].dims['row']
if args.output_type is not None:
output_type = np.dtype(args.output_type)
else:
output_type = np.result_type(np.dtype(args.real_type), np.complex64)
data_bytes = output_type.itemsize
bytes_per_row = max_chan_chunk * ncorr * data_bytes
memory_per_row = bytes_per_row # model
memory_per_row += 3 * 8 # uvw
if mstype == 'zarr':
if args.model_column in xds[0].keys():
model_chunks = getattr(xds[0], args.model_column).data.chunks
else:
model_chunks = xds[0].DATA.data.chunks
print('Chunking model same as data')
# get approx image size
# this is not a conservative estimate when multiple SPW's map to a single
# imaging band
pixel_bytes = np.dtype(args.output_type).itemsize
band_size = nx * ny * pixel_bytes
if args.host_address is None:
# full image on single node
row_chunk = plan_row_chunk(mem_limit / nworkers, band_size, nrow,
memory_per_row, nthreads_per_worker)
else:
# single band per node
row_chunk = plan_row_chunk(mem_limit, band_size, nrow, memory_per_row,
nthreads_per_worker)
if args.row_chunks is not None:
row_chunk = int(args.row_chunks)
if row_chunk == -1:
row_chunk = nrow
print(
"nrows = %i, row chunks set to %i for a total of %i chunks per node" %
(nrow, row_chunk, int(np.ceil(nrow / row_chunk))),
file=log)
chunks = {}
for ims in ms:
chunks[ims] = [] # xds_from_ms expects a list per ds
for spw in freqs[ims]:
chunks[ims].append({
'row': row_chunk,
'chan': chan_chunks[ims][spw]['chan']
})
model = da.from_array(model.astype(args.real_type),
chunks=(1, nx, ny),
name=False)
writes = []
radec = None # assumes we are only imaging field 0 of first MS
for ims in ms:
xds = xds_from_ms(ims, chunks=chunks[ims], columns=('UVW'))
# subtables
ddids = xds_from_table(ims + "::DATA_DESCRIPTION")
fields = xds_from_table(ims + "::FIELD")
spws = xds_from_table(ims + "::SPECTRAL_WINDOW")
pols = xds_from_table(ims + "::POLARIZATION")
# subtable data
ddids = dask.compute(ddids)[0]
fields = dask.compute(fields)[0]
spws = dask.compute(spws)[0]
pols = dask.compute(pols)[0]
out_data = []
for ds in xds:
field = fields[ds.FIELD_ID]
radec = field.PHASE_DIR.data.squeeze()
# check fields match
if radec is None:
radec = field.PHASE_DIR.data.squeeze()
if not np.array_equal(radec, field.PHASE_DIR.data.squeeze()):
continue
spw = ds.DATA_DESC_ID # this is not correct, need to use spw
uvw = clone(ds.UVW.data)
bands = band_mapping[ims][spw]
model = model[list(bands), :, :]
vis = im2vis(uvw,
freqs[ims][spw],
model,
freq_bin_idx[ims][spw],
freq_bin_counts[ims][spw],
cell_rad,
nthreads=ngridder_threads,
epsilon=args.epsilon,
do_wstacking=args.wstack)
model_vis = restore_corrs(vis, ncorr)
if mstype == 'zarr':
model_vis = model_vis.rechunk(model_chunks)
uvw = uvw.rechunk((model_chunks[0], 3))
out_ds = ds.assign(
**{
args.model_column: (("row", "chan", "corr"), model_vis),
'UVW': (("row", "three"), uvw)
})
# out_ds = ds.assign(**{args.model_column: (("row", "chan", "corr"), model_vis)})
out_data.append(out_ds)
writes.append(xds_to_table(out_data, ims, columns=[args.model_column]))
dask.visualize(*writes,
filename=args.output_filename + '_predict_graph.pdf',
optimize_graph=False,
collapse_outputs=True)
if not args.mock:
with performance_report(filename=args.output_filename +
'_predict_per.html'):
dask.compute(writes, optimize_graph=False)
print("All done here.", file=log)
def psf(**kw):
'''
Create a psf image from a list of measurement setsand write out the
Mueller weights.
The psf image cube is not normalised by wsum as this destroyes
information. The MFS image is written out in units of Jy/beam
and should have a peak of one otherwise something has gone wrong.
The --field-of-view and --super-resolution-factor options
(equivalently --cell-size, --nx and --ny) pertain to the size of
the image (eg. dirty and model). The size of the PSF output image
is controlled by the --psf-oversize option.
The Stokes I weights required to apply the Hessian are also written out
to a zarr data set called output-filename.zarr. This data set does not
adhere to the MSv2 specs and is only meant to be used to apply the
Hessian. In particular, the weights written out are a combination of
imaging weights and the "Mueller" weights.
If a host address is provided the computation can be distributed
over imaging band and row. When using a distributed scheduler both
mem-limit and nthreads is per node and have to be specified.
When using a local cluster, mem-limit and nthreads refer to the global
memory and threads available, respectively. By default the gridder will
use all available resources.
Disclaimer - Memory budgeting is still very crude!
On a local cluster, the default is to use:
nworkers = nband
nthreads-per-worker = 1
They have to be specified in ~.config/dask/jobqueue.yaml in the
distributed case.
if LocalCluster:
ngridder-threads = nthreads//(nworkers*nthreads_per_worker)
else:
ngridder-threads = nthreads//nthreads-per-worker
'''
args = OmegaConf.create(kw)
pyscilog.log_to_file(args.output_filename + '.log')
from glob import glob
ms = glob(args.ms)
try:
assert len(ms) > 0
args.ms = ms
except:
raise ValueError(f"No MS at {args.ms}")
if args.nworkers is None:
args.nworkers = args.nband
OmegaConf.set_struct(args, True)
with ExitStack() as stack:
from pfb import set_client
args = set_client(args, stack, log)
# TODO - prettier config printing
print('Input Options:', file=log)
for key in args.keys():
print(' %25s = %s' % (key, args[key]), file=log)
return _psf(**args)
def _restore(stack, **kw):
args = OmegaConf.create(kw)
OmegaConf.set_struct(args, True)
pyscilog.log_to_file(args.output_filename + '.log')
pyscilog.enable_memory_logging(level=3)
# number of threads per worker
if args.nthreads is None:
if args.host_address is not None:
raise ValueError(
"You have to specify nthreads when using a distributed scheduler"
)
import multiprocessing
nthreads = multiprocessing.cpu_count()
args.nthreads = nthreads
else:
nthreads = args.nthreads
# configure memory limit
if args.mem_limit is None:
if args.host_address is not None:
raise ValueError(
"You have to specify mem-limit when using a distributed scheduler"
)
import psutil
mem_limit = int(psutil.virtual_memory()[0] /
1e9) # 100% of memory by default
args.mem_limit = mem_limit
else:
mem_limit = args.mem_limit
nband = args.nband
if args.nworkers is None:
nworkers = nband
args.nworkers = nworkers
else:
nworkers = args.nworkers
if args.nthreads_per_worker is None:
nthreads_per_worker = 1
args.nthreads_per_worker = nthreads_per_worker
else:
nthreads_per_worker = args.nthreads_per_worker
# the number of chunks being read in simultaneously is equal to
# the number of dask threads
nthreads_dask = nworkers * nthreads_per_worker
if args.ngridder_threads is None:
if args.host_address is not None:
ngridder_threads = nthreads // nthreads_per_worker
else:
ngridder_threads = nthreads // nthreads_dask
args.ngridder_threads = ngridder_threads
else:
ngridder_threads = args.ngridder_threads
ms = list(ms)
print('Input Options:', file=log)
for key in kw.keys():
print(' %25s = %s' % (key, args[key]), file=log)
# numpy imports have to happen after this step
from pfb import set_client
set_client(nthreads, mem_limit, nworkers, nthreads_per_worker,
args.host_address, stack, log)
import numpy as np
from astropy.io import fits
mhdr = fits.getheader(args.model)
from pfb.utils.fits import load_fits
model = load_fits(args.model).squeeze() # drop Stokes axis
# check images compatible
rhdr = fits.getheader(args.residual)
from pfb.utils.fits import compare_headers
compare_headers(mhdr, rhdr)
residual = load_fits(args.residual).squeeze()
# fit restoring psf
from pfb.utils.misc import fitcleanbeam
psf = load_fits(args.psf, dtype=args.real_type).squeeze()
nband, nx_psf, ny_psf = psf.shape
wsums = np.amax(psf.reshape(args.nband, nx_psf, ny_psf), axis=1)
wsum = np.sum(wsums)
psf /= wsum
psf_mfs = np.sum(psf, axis=0)
# fit restoring psf
GaussPar = fitcleanbeam(psf_mfs[None], level=0.5, pixsize=1.0)
GaussPars = fitcleanbeam(psf, level=0.5, pixsize=1.0)
cpsf_mfs = np.zeros(psf_mfs.shape, dtype=args.real_type)
cpsf = np.zeros(psf.shape, dtype=args.real_type)
lpsf = np.arange(-R.nx_psf / 2, R.nx_psf / 2)
mpsf = np.arange(-R.ny_psf / 2, R.ny_psf / 2)
xx, yy = np.meshgrid(lpsf, mpsf, indexing='ij')
cpsf_mfs = Gaussian2D(xx, yy, GaussPar[0], normalise=False)
for v in range(args.nband):
cpsf[v] = Gaussian2D(xx, yy, GaussPars[v], normalise=False)
from pfb.utils.fits import add_beampars
GaussPar = list(GaussPar[0])
GaussPar[0] *= args.cell_size / 3600
GaussPar[1] *= args.cell_size / 3600
GaussPar = tuple(GaussPar)
hdr_psf_mfs = add_beampars(hdr_psf_mfs, GaussPar)
save_fits(args.outfile + '_cpsf_mfs.fits', cpsf_mfs, hdr_psf_mfs)
save_fits(args.outfile + '_psf_mfs.fits', psf_mfs, hdr_psf_mfs)
if args.beam is not None:
bhdr = fits.getheader(args.beam)
compare_headers(mhdr, bhdr)
beam = load_fits(args.beam).squeeze()
model = np.where(beam > args.pb_min, model / beam, 0.0)
nband, nx, ny = model.shape
guassparf = ()
if nband > 1:
for b in range(nband):
guassparf += (rhdr['BMAJ' + str(b)], rhdr['BMIN' + str(b)],
rhdr['BPA' + str(b)])
else:
guassparf += (rhdr['BMAJ'], rhdr['BMIN'], rhdr['BPA'])
# if args.convolve_residuals:
cellx = np.abs(mhdr['CDELT1'])
celly = np.abs(mhdr['CDELT2'])
from pfb.utils.restoration import restore_image
def spifit(**kw):
"""
Spectral index fitter
"""
args = OmegaConf.create(kw)
pyscilog.log_to_file(args.output_filename + '.log')
from glob import glob
from omegaconf import ListConfig
# image is either a string or a list of strings that we want to glob on
if isinstance(args.image, str):
image = sorted(glob(args.image))
elif isinstance(args.image, list) or isinstance(args.image, ListConfig):
image = []
for i in len(args.image):
image.append(sorted(glob(args.image[i])))
# make sure it's not empty
try:
assert len(image) > 0
args.image = image
except:
raise ValueError(f"No image at {args.image}")
# same goes for the residual except that it may also be None
if isinstance(args.residual, str):
residual = sorted(glob(args.residual))
elif isinstance(args.residual, list) or isinstance(args.residual, ListConfig):
residual = []
for i in len(args.residual):
residual.append(sorted(glob(args.residual[i])))
if args.residual is not None:
try:
assert len(residual) > 0
args.residual = residual
except:
raise ValueError(f"No residual at {args.residual}")
# we also need the same number of residuals as images
try:
assert len(args.image) == len(args.residual)
except:
raise ValueError(f"Number of images and residuals need to "
"match")
else:
print("No residual passed in!", file=log)
# and finally the beam model
if isinstance(args.beam_model, str):
beam_model = sorted(glob(args.beam_model))
elif isinstance(args.beam_model, list) or isinstance(args.beam_model, ListConfig):
beam_model = []
for i in len(args.beam_model):
beam_model.append(sorted(glob(args.beam_model[i])))
if args.beam_model is not None:
try:
assert len(beam_model) > 0
args.beam_model = beam_model
except:
raise ValueError(f"No beam model at {args.beam_model}")
try:
assert len(args.image) == len(args.beam_model)
except:
raise ValueError(f"Number of images and beam models need to "
"match")
else:
print("Not doing any form of primary beam correction", file=log)
# LB - TODO: can we sort them along freq at this point already?
OmegaConf.set_struct(args, True)
with ExitStack() as stack:
from pfb import set_client
args = set_client(args, stack, log)
# TODO - prettier config printing
print('Input Options:', file=log)
for key in args.keys():
print(' %25s = %s' % (key, args[key]), file=log)
return _spifit(**args)
def binterp(**kw):
"""
Beam interpolator
Interpolate beams and stack cubes one MS and one spectral window at a time.
"""
args = OmegaConf.create(kw)
from glob import glob
image = sorted(glob(args.image))
try:
assert len(image) > 0
args.image = image
except:
raise ValueError(f"No image at {args.image}")
if args.output_dir is None:
args.output_dir = os.path.dirname(args.image[0])
pyscilog.log_to_file(args.output_dir + args.postfix.strip('fits') + 'log')
if args.ms is not None:
ms = glob(args.ms)
try:
assert len(ms) == 1
args.ms = ms[0]
except:
raise ValueError(
f"There must be exactly one MS matching {args.ms} if provided")
if not isinstance(args.beam_model, str):
raise ValueError("Only string beam patterns allowed")
else:
# we are either using JimBeam or globbing for beam patterns
if args.beam_model.lower() == 'jimbeam':
args.beam_model = args.beam_model.lower()
band = args.band.lower()
if band != 'l' and band != 'uhf':
raise ValueError("Only l or uhf band supported with "
"JimBeam")
else:
print("Using %s band beam model" % args.band, file=log)
elif args.beam_model.lower().endswith('.fits'):
beam_model = glob(args.beam_model)
try:
assert len(beam_model) > 0
except:
raise ValueError(f"No beam model at {args.beam_model}")
else:
raise ValueError("Unknown beam model provided. "
"Either use JimBeam or pass in the fits beam "
"patterns")
OmegaConf.set_struct(args, True)
with ExitStack() as stack:
from pfb import set_client
args = set_client(args, stack, log)
# TODO - prettier config printing
print('Input Options:', file=log)
for key in args.keys():
print(' %25s = %s' % (key, args[key]), file=log)
return _binterp(**args)
def _residual(ms, stack, **kw):
args = OmegaConf.create(kw)
OmegaConf.set_struct(args, True)
pyscilog.log_to_file(args.output_filename + '.log')
pyscilog.enable_memory_logging(level=3)
# number of threads per worker
if args.nthreads is None:
if args.host_address is not None:
raise ValueError(
"You have to specify nthreads when using a distributed scheduler"
)
import multiprocessing
nthreads = multiprocessing.cpu_count()
args.nthreads = nthreads
else:
nthreads = args.nthreads
# configure memory limit
if args.mem_limit is None:
if args.host_address is not None:
raise ValueError(
"You have to specify mem-limit when using a distributed scheduler"
)
import psutil
mem_limit = int(psutil.virtual_memory()[0] /
1e9) # 100% of memory by default
args.mem_limit = mem_limit
else:
mem_limit = args.mem_limit
nband = args.nband
if args.nworkers is None:
nworkers = nband
args.nworkers = nworkers
else:
nworkers = args.nworkers
if args.nthreads_per_worker is None:
nthreads_per_worker = 1
args.nthreads_per_worker = nthreads_per_worker
else:
nthreads_per_worker = args.nthreads_per_worker
# the number of chunks being read in simultaneously is equal to
# the number of dask threads
nthreads_dask = nworkers * nthreads_per_worker
if args.ngridder_threads is None:
if args.host_address is not None:
ngridder_threads = nthreads // nthreads_per_worker
else:
ngridder_threads = nthreads // nthreads_dask
args.ngridder_threads = ngridder_threads
else:
ngridder_threads = args.ngridder_threads
ms = list(ms)
print('Input Options:', file=log)
for key in kw.keys():
print(' %25s = %s' % (key, args[key]), file=log)
# numpy imports have to happen after this step
from pfb import set_client
set_client(nthreads, mem_limit, nworkers, nthreads_per_worker,
args.host_address, stack, log)
import numpy as np
from pfb.utils.misc import chan_to_band_mapping
import dask
from dask.graph_manipulation import clone
from dask.distributed import performance_report
from daskms import xds_from_storage_ms as xds_from_ms
from daskms import xds_from_storage_table as xds_from_table
import dask.array as da
from africanus.constants import c as lightspeed
from africanus.gridding.wgridder.dask import residual as im2residim
from ducc0.fft import good_size
from pfb.utils.misc import stitch_images, plan_row_chunk
from pfb.utils.fits import set_wcs, save_fits
# chan <-> band mapping
freqs, freq_bin_idx, freq_bin_counts, freq_out, band_mapping, chan_chunks = chan_to_band_mapping(
ms, nband=nband)
# gridder memory budget
max_chan_chunk = 0
max_freq = 0
for ims in ms:
for spw in freqs[ims]:
counts = freq_bin_counts[ims][spw].compute()
freq = freqs[ims][spw].compute()
max_chan_chunk = np.maximum(max_chan_chunk, counts.max())
max_freq = np.maximum(max_freq, freq.max())
# assumes measurement sets have the same columns,
# number of correlations etc.
xds = xds_from_ms(ms[0])
ncorr = xds[0].dims['corr']
nrow = xds[0].dims['row']
data_bytes = getattr(xds[0], args.data_column).data.itemsize
bytes_per_row = max_chan_chunk * ncorr * data_bytes
memory_per_row = bytes_per_row
# real valued weights
wdims = getattr(xds[0], args.weight_column).data.ndim
if wdims == 2: # WEIGHT
memory_per_row += ncorr * data_bytes / 2
else: # WEIGHT_SPECTRUM
memory_per_row += bytes_per_row / 2
# flags (uint8 or bool)
memory_per_row += np.dtype(np.uint8).itemsize * max_chan_chunk * ncorr
# UVW
memory_per_row += xds[0].UVW.data.itemsize * 3
# ANTENNA1/2
memory_per_row += xds[0].ANTENNA1.data.itemsize * 2
columns = (args.data_column, args.weight_column, args.flag_column, 'UVW',
'ANTENNA1', 'ANTENNA2')
# flag row
if 'FLAG_ROW' in xds[0]:
columns += ('FLAG_ROW', )
memory_per_row += xds[0].FLAG_ROW.data.itemsize
# imaging weights
if args.imaging_weight_column is not None:
columns += (args.imaging_weight_column, )
memory_per_row += bytes_per_row / 2
# Mueller term (complex valued)
if args.mueller_column is not None:
columns += (args.mueller_column, )
memory_per_row += bytes_per_row
# get max uv coords over all fields
uvw = []
u_max = 0.0
v_max = 0.0
for ims in ms:
xds = xds_from_ms(ims, columns=('UVW'), chunks={'row': -1})
for ds in xds:
uvw = ds.UVW.data
u_max = da.maximum(u_max, abs(uvw[:, 0]).max())
v_max = da.maximum(v_max, abs(uvw[:, 1]).max())
uv_max = da.maximum(u_max, v_max)
uv_max = uv_max.compute()
del uvw
# image size
cell_N = 1.0 / (2 * uv_max * max_freq / lightspeed)
if args.cell_size is not None:
cell_size = args.cell_size
cell_rad = cell_size * np.pi / 60 / 60 / 180
if cell_N / cell_rad < 1:
raise ValueError(
"Requested cell size too small. "
"Super resolution factor = ", cell_N / cell_rad)
print("Super resolution factor = %f" % (cell_N / cell_rad), file=log)
else:
cell_rad = cell_N / args.super_resolution_factor
cell_size = cell_rad * 60 * 60 * 180 / np.pi
print("Cell size set to %5.5e arcseconds" % cell_size, file=log)
if args.nx is None:
fov = args.field_of_view * 3600
npix = int(fov / cell_size)
if npix % 2:
npix += 1
nx = good_size(npix)
ny = good_size(npix)
else:
nx = args.nx
ny = args.ny if args.ny is not None else nx
print("Image size set to (%i, %i, %i)" % (nband, nx, ny), file=log)
# get approx image size
# this is not a conservative estimate when multiple SPW's map to a single
# imaging band
pixel_bytes = np.dtype(args.output_type).itemsize
band_size = nx * ny * pixel_bytes
if args.host_address is None:
# full image on single node
row_chunk = plan_row_chunk(mem_limit / nworkers, band_size, nrow,
memory_per_row, nthreads_per_worker)
else:
# single band per node
row_chunk = plan_row_chunk(mem_limit, band_size, nrow, memory_per_row,
nthreads_per_worker)
if args.row_chunks is not None:
row_chunk = int(args.row_chunks)
if row_chunk == -1:
row_chunk = nrow
print(
"nrows = %i, row chunks set to %i for a total of %i chunks per node" %
(nrow, row_chunk, int(np.ceil(nrow / row_chunk))),
file=log)
chunks = {}
for ims in ms:
chunks[ims] = [] # xds_from_ms expects a list per ds
for spw in freqs[ims]:
chunks[ims].append({
'row': row_chunk,
'chan': chan_chunks[ims][spw]['chan']
})
dirties = []
radec = None # assumes we are only imaging field 0 of first MS
for ims in ms:
xds = xds_from_ms(ims, chunks=chunks[ims], columns=columns)
# subtables
ddids = xds_from_table(ims + "::DATA_DESCRIPTION")
fields = xds_from_table(ims + "::FIELD")
spws = xds_from_table(ims + "::SPECTRAL_WINDOW")
pols = xds_from_table(ims + "::POLARIZATION")
# subtable data
ddids = dask.compute(ddids)[0]
fields = dask.compute(fields)[0]
spws = dask.compute(spws)[0]
pols = dask.compute(pols)[0]
for ds in xds:
field = fields[ds.FIELD_ID]
# check fields match
if radec is None:
radec = field.PHASE_DIR.data.squeeze()
if not np.array_equal(radec, field.PHASE_DIR.data.squeeze()):
continue
# this is not correct, need to use spw
spw = ds.DATA_DESC_ID
uvw = clone(ds.UVW.data)
data = getattr(ds, args.data_column).data
dataxx = data[:, :, 0]
datayy = data[:, :, -1]
weights = getattr(ds, args.weight_column).data
if len(weights.shape) < 3:
weights = da.broadcast_to(weights[:, None, :],
data.shape,
chunks=data.chunks)
if args.imaging_weight_column is not None:
imaging_weights = getattr(ds, args.imaging_weight_column).data
if len(imaging_weights.shape) < 3:
imaging_weights = da.broadcast_to(imaging_weights[:,
None, :],
data.shape,
chunks=data.chunks)
weightsxx = imaging_weights[:, :, 0] * weights[:, :, 0]
weightsyy = imaging_weights[:, :, -1] * weights[:, :, -1]
else:
weightsxx = weights[:, :, 0]
weightsyy = weights[:, :, -1]
# apply adjoint of mueller term.
# Phases modify data amplitudes modify weights.
if args.mueller_column is not None:
mueller = getattr(ds, args.mueller_column).data
dataxx *= da.exp(-1j * da.angle(mueller[:, :, 0]))
datayy *= da.exp(-1j * da.angle(mueller[:, :, -1]))
weightsxx *= da.absolute(mueller[:, :, 0])
weightsyy *= da.absolute(mueller[:, :, -1])
# weighted sum corr to Stokes I
weights = weightsxx + weightsyy
data = (weightsxx * dataxx + weightsyy * datayy)
# TODO - turn off this stupid warning
data = da.where(weights, data / weights, 0.0j)
# MS may contain auto-correlations
if 'FLAG_ROW' in xds[0]:
frow = ds.FLAG_ROW.data | (ds.ANTENNA1.data
== ds.ANTENNA2.data)
else:
frow = (ds.ANTENNA1.data == ds.ANTENNA2.data)
# only keep data where both corrs are unflagged
flag = getattr(ds, args.flag_column).data
flagxx = flag[:, :, 0]
flagyy = flag[:, :, -1]
# ducc0 uses uint8 mask not flag
mask = ~da.logical_or((flagxx | flagyy), frow[:, None])
dirty = vis2im(uvw,
freqs[ims][spw],
data,
freq_bin_idx[ims][spw],
freq_bin_counts[ims][spw],
nx,
ny,
cell_rad,
weights=weights,
flag=mask.astype(np.uint8),
nthreads=ngridder_threads,
epsilon=args.epsilon,
do_wstacking=args.wstack,
double_accum=args.double_accum)
dirties.append(dirty)
# dask.visualize(dirties, filename=args.output_filename + '_graph.pdf', optimize_graph=False)
if not args.mock:
# result = dask.compute(dirties, wsum, optimize_graph=False)
with performance_report(filename=args.output_filename + '_per.html'):
result = dask.compute(dirties, optimize_graph=False)
dirties = result[0]
dirty = stitch_images(dirties, nband, band_mapping)
hdr = set_wcs(cell_size / 3600, cell_size / 3600, nx, ny, radec,
freq_out)
save_fits(args.output_filename + '_dirty.fits',
dirty,
hdr,
dtype=args.output_type)
print("All done here.", file=log)