示例#1
0
def grid_gaintable_to_screen(vis,
                             gaintables,
                             screen,
                             height=3e5,
                             gaintable_slices=None,
                             scale=1.0,
                             r0=5e3,
                             type_atmosphere='ionosphere',
                             vis_slices=None,
                             **kwargs):
    """ Grid a gaintable to a screen image

    Screen axes are ['XX', 'YY', 'TIME', 'FREQ']

    The phases are just averaged per grid cell, no phase unwrapping is performed.

    :param vis:
    :param gaintables: input gaintables
    :param screen:
    :param height: Height (in m) of screen above telescope e.g. 3e5
    :param r0: r0 in meters
    :param type_atmosphere: 'ionosphere' or 'troposphere'
    :param scale: Multiply the screen by this factor
    :return: gridded screen image, weights image
    """
    assert isinstance(vis, BlockVisibility)

    station_locations = vis.configuration.xyz

    nant = station_locations.shape[0]
    t2r = numpy.pi / 43200.0

    newscreen = create_empty_image_like(screen)
    weights = create_empty_image_like(screen)
    nchan, ntimes, ny, nx = screen.shape

    number_no_weight = 0

    for gaintable in gaintables:
        ha_zero = numpy.average(calculate_blockvisibility_hourangles(vis))
        for iha, rows in enumerate(
                vis_timeslice_iter(vis, vis_slices=vis_slices)):
            v = create_visibility_from_rows(vis, rows)
            ha = numpy.average(
                calculate_blockvisibility_hourangles(v) -
                ha_zero).to('rad').value
            pp = find_pierce_points(station_locations,
                                    (gaintable.phasecentre.ra.rad + t2r * ha) *
                                    units.rad,
                                    gaintable.phasecentre.dec,
                                    height=height,
                                    phasecentre=vis.phasecentre)

            scr = numpy.angle(gaintable.gain[0, :, 0, 0, 0])
            wt = gaintable.weight[0, :, 0, 0, 0]
            for ant in range(nant):
                pp0 = pp[ant][0:2]
                for freq in vis.frequency:
                    scale = numpy.power(r0 / 5000.0, -5.0 / 3.0)
                    if type_atmosphere == 'troposphere':
                        # In troposphere files, the units are phase in radians.
                        screen_to_phase = scale
                    else:
                        # In the ionosphere file, the units are dTEC.
                        screen_to_phase = -scale * 8.44797245e9 / freq
                    worldloc = [pp0[0], pp0[1], ha, freq]
                    pixloc = newscreen.wcs.wcs_world2pix([worldloc],
                                                         0)[0].astype('int')
                    assert pixloc[0] >= 0
                    assert pixloc[0] < nx
                    assert pixloc[1] >= 0
                    assert pixloc[1] < ny
                    pixloc[3] = 0
                    newscreen.data[
                        pixloc[3], pixloc[2], pixloc[1],
                        pixloc[0]] += wt[ant] * scr[ant] / screen_to_phase
                    weights.data[pixloc[3], pixloc[2], pixloc[1],
                                 pixloc[0]] += wt[ant]
                    if wt[ant] == 0.0:
                        number_no_weight += 1
    if number_no_weight > 0:
        log.warning(
            "grid_gaintable_to_screen: %d pierce points are have no weight" %
            (number_no_weight))

    assert numpy.max(weights.data) > 0.0, "No points were gridded"

    newscreen.data[weights.data > 0.0] = newscreen.data[
        weights.data > 0.0] / weights.data[weights.data > 0.0]

    return newscreen, weights
示例#2
0
def create_gaintable_from_screen(vis,
                                 sc,
                                 screen,
                                 height=3e5,
                                 vis_slices=None,
                                 scale=1.0,
                                 r0=5e3,
                                 type_atmosphere='ionosphere',
                                 **kwargs):
    """ Create gaintables from a screen calculated using ARatmospy

    Screen axes are ['XX', 'YY', 'TIME', 'FREQ']

    :param vis:
    :param sc: Sky components for which pierce points are needed
    :param screen:
    :param height: Height (in m) of screen above telescope e.g. 3e5
    :param r0: r0 in meters
    :param type_atmosphere: 'ionosphere' or 'troposphere'
    :param scale: Multiply the screen by this factor
    :return:
    """
    assert isinstance(vis, BlockVisibility)

    station_locations = vis.configuration.xyz

    scale = numpy.power(r0 / 5000.0, -5.0 / 3.0)
    if type_atmosphere == 'troposphere':
        # In troposphere files, the units are phase in radians.
        screen_to_phase = scale
    else:
        # In the ionosphere file, the units are dTEC.
        screen_to_phase = -scale * 8.44797245e9 / numpy.array(vis.frequency)

    nant = station_locations.shape[0]
    t2r = numpy.pi / 43200.0
    gaintables = [
        create_gaintable_from_blockvisibility(vis, **kwargs) for i in sc
    ]

    number_bad = 0
    number_good = 0

    ha_zero = numpy.average(calculate_blockvisibility_hourangles(vis))
    for iha, rows in enumerate(vis_timeslice_iter(vis, vis_slices=vis_slices)):
        v = create_visibility_from_rows(vis, rows)
        ha = numpy.average(calculate_blockvisibility_hourangles(v) -
                           ha_zero).to('rad').value
        for icomp, comp in enumerate(sc):
            pp = find_pierce_points(station_locations,
                                    (comp.direction.ra.rad + t2r * ha) *
                                    units.rad,
                                    comp.direction.dec,
                                    height=height,
                                    phasecentre=vis.phasecentre)
            scr = numpy.zeros([nant])
            for ant in range(nant):
                pp0 = pp[ant][0:2]
                # Using narrow band approach - we should loop over frequency
                try:
                    worldloc = [
                        pp0[0], pp0[1], ha,
                        numpy.average(vis.frequency)
                    ]
                    pixloc = screen.wcs.wcs_world2pix([worldloc],
                                                      0)[0].astype('int')
                    if type_atmosphere == 'troposphere':
                        pixloc[3] = 0
                    scr[ant] = screen_to_phase * screen.data[
                        pixloc[3], pixloc[2], pixloc[1], pixloc[0]]
                    number_good += 1
                except (ValueError, IndexError):
                    number_bad += 1
                    scr[ant] = 0.0

            # axes of gaintable.gain are time, ant, nchan, nrec
            gaintables[icomp].gain[iha, :, :, :] = numpy.exp(
                1j * scr)[..., numpy.newaxis, numpy.newaxis, numpy.newaxis]
            gaintables[icomp].phasecentre = comp.direction

    assert number_good > 0, "create_gaintable_from_screen: There are no pierce points inside the atmospheric screen image"
    if number_bad > 0:
        log.warning(
            "create_gaintable_from_screen: %d pierce points are inside the atmospheric screen image"
            % (number_good))
        log.warning(
            "create_gaintable_from_screen: %d pierce points are outside the atmospheric screen image"
            % (number_bad))

    return gaintables
示例#3
0
def predict_list_serial_workflow(vis_list,
                                 model_imagelist,
                                 context,
                                 vis_slices=1,
                                 facets=1,
                                 gcfcf=None,
                                 **kwargs):
    """Predict, iterating over both the scattered vis_list and image

    The visibility and image are scattered, the visibility is predicted on each part, and then the
    parts are assembled.

    :param vis_list: list of vis
    :param model_imagelist: Model used to determine image parameters
    :param vis_slices: Number of vis slices (w stack or timeslice)
    :param facets: Number of facets (per axis)
    :param context: Type of processing e.g. 2d, wstack, timeslice or facets
    :param gcfcg: tuple containing grid correction and convolution function
    :param kwargs: Parameters for functions in components
    :return: List of vis_lists
   """

    assert len(vis_list) == len(
        model_imagelist), "Model must be the same length as the vis_list"

    # Predict_2d does not clear the vis so we have to do it here.
    vis_list = zero_list_serial_workflow(vis_list)

    c = imaging_context(context)
    vis_iter = c['vis_iterator']
    predict = c['predict']

    if facets % 2 == 0 or facets == 1:
        actual_number_facets = facets
    else:
        actual_number_facets = facets - 1

    def predict_ignore_none(vis, model, g):
        if vis is not None:
            assert isinstance(vis, Visibility) or isinstance(
                vis, BlockVisibility), vis
            assert isinstance(model, Image), model
            return predict(vis, model, context=context, gcfcf=g, **kwargs)
        else:
            return None

    if gcfcf is None:
        gcfcf = [create_pswf_convolutionfunction(m) for m in model_imagelist]

    # Loop over all frequency windows
    if facets == 1:
        image_results_list = list()
        for ivis, sub_vis_list in enumerate(vis_list):
            if len(gcfcf) > 1:
                g = gcfcf[ivis]
            else:
                g = gcfcf[0]
            # Loop over sub visibility
            vis_predicted = copy_visibility(sub_vis_list, zero=True)
            for rows in vis_iter(sub_vis_list, vis_slices):
                row_vis = create_visibility_from_rows(sub_vis_list, rows)
                row_vis_predicted = predict_ignore_none(
                    row_vis, model_imagelist[ivis], g)
                if row_vis_predicted is not None:
                    vis_predicted.data['vis'][
                        rows, ...] = row_vis_predicted.data['vis']
            image_results_list.append(vis_predicted)

        return image_results_list
    else:
        image_results_list = list()
        for ivis, sub_vis_list in enumerate(vis_list):
            # Create the graph to divide an image into facets. This is by reference.
            facet_lists = image_scatter_facets(model_imagelist[ivis],
                                               facets=facets)
            facet_vis_lists = list()
            sub_vis_lists = visibility_scatter(sub_vis_list, vis_iter,
                                               vis_slices)

            # Loop over sub visibility
            for sub_sub_vis_list in sub_vis_lists:
                facet_vis_results = list()
                # Loop over facets
                for facet_list in facet_lists:
                    # Predict visibility for this subvisibility from this facet
                    facet_vis_list = predict_ignore_none(
                        sub_sub_vis_list, facet_list, None)
                    facet_vis_results.append(facet_vis_list)
                # Sum the current sub-visibility over all facets
                facet_vis_lists.append(sum_predict_results(facet_vis_results))
            # Sum all sub-visibilties
            image_results_list.append(
                visibility_gather(facet_vis_lists, sub_vis_list, vis_iter))
        return image_results_list
示例#4
0
def invert_list_serial_workflow(vis_list,
                                template_model_imagelist,
                                dopsf=False,
                                normalize=True,
                                facets=1,
                                vis_slices=1,
                                context='2d',
                                gcfcf=None,
                                **kwargs):
    """ Sum results from invert, iterating over the scattered image and vis_list

    :param vis_list: list of vis
    :param template_model_imagelist: list of template models
    :param dopsf: Make the PSF instead of the dirty image
    :param facets: Number of facets
    :param normalize: Normalize by sumwt
    :param vis_slices: Number of slices
    :param context: Imaging context
    :param gcfcg: tuple containing grid correction and convolution function
    :param kwargs: Parameters for functions in components
    :return: List of (image, sumwt) tuples, one per vis in vis_list

    For example::

        model_list = [create_image_from_visibility
            (v, npixel=npixel, cellsize=cellsize, polarisation_frame=pol_frame)
            for v in vis_list]

        dirty_list = invert_list_serial_workflow(vis_list, template_model_imagelist=model_list, context='wstack',
                                                    vis_slices=51)
        dirty, sumwt = dirty_list[centre]

   """

    if not isinstance(template_model_imagelist, collections.abc.Iterable):
        template_model_imagelist = [template_model_imagelist]

    c = imaging_context(context)
    vis_iter = c['vis_iterator']
    invert = c['invert']

    if facets % 2 == 0 or facets == 1:
        actual_number_facets = facets
    else:
        actual_number_facets = max(1, (facets - 1))

    def gather_image_iteration_results(results, template_model):
        result = create_empty_image_like(template_model)
        i = 0
        sumwt = numpy.zeros([template_model.nchan, template_model.npol])
        for dpatch in image_scatter_facets(result, facets=facets):
            assert i < len(
                results), "Too few results in gather_image_iteration_results"
            if results[i] is not None:
                assert len(results[i]) == 2, results[i]
                dpatch.data[...] = results[i][0].data[...]
                sumwt += results[i][1]
                i += 1
        return result, sumwt

    def invert_ignore_none(vis, model, gg):
        if vis is not None:

            return invert(vis,
                          model,
                          context=context,
                          dopsf=dopsf,
                          normalize=normalize,
                          gcfcf=gg,
                          **kwargs)
        else:
            return create_empty_image_like(model), numpy.zeros(
                [model.nchan, model.npol])

    # If we are doing facets, we need to create the gcf for each image
    if gcfcf is None and facets == 1:
        gcfcf = [create_pswf_convolutionfunction(template_model_imagelist[0])]

    # Loop over all vis_lists independently
    results_vislist = list()
    if facets == 1:
        for ivis, sub_vis_list in enumerate(vis_list):
            if len(gcfcf) > 1:
                g = gcfcf[ivis]
            else:
                g = gcfcf[0]
            # Iterate within each vis_list
            result_image = create_empty_image_like(
                template_model_imagelist[ivis])
            result_sumwt = numpy.zeros([
                template_model_imagelist[ivis].nchan,
                template_model_imagelist[ivis].npol
            ])
            for rows in vis_iter(sub_vis_list, vis_slices):
                row_vis = create_visibility_from_rows(sub_vis_list, rows)
                result = invert_ignore_none(row_vis,
                                            template_model_imagelist[ivis], g)
                if result is not None:
                    result_image.data += result[1][:, :, numpy.newaxis, numpy.
                                                   newaxis] * result[0].data
                    result_sumwt += result[1]
            result_image = normalize_sumwt(result_image, result_sumwt)
            results_vislist.append((result_image, result_sumwt))
    else:
        for ivis, sub_vis_list in enumerate(vis_list):
            # Create the graph to divide an image into facets. This is by reference.
            facet_lists = image_scatter_facets(template_model_imagelist[ivis],
                                               facets=facets)
            # Create the graph to divide the visibility into slices. This is by copy.
            sub_sub_vis_lists = visibility_scatter(sub_vis_list,
                                                   vis_iter,
                                                   vis_slices=vis_slices)

            # Iterate within each vis_list
            vis_results = list()
            for sub_sub_vis_list in sub_sub_vis_lists:
                facet_vis_results = list()
                for facet_list in facet_lists:
                    facet_vis_results.append(
                        invert_ignore_none(sub_sub_vis_list, facet_list, None))
                vis_results.append(
                    gather_image_iteration_results(
                        facet_vis_results, template_model_imagelist[ivis]))
            results_vislist.append(sum_invert_results(vis_results))

    return results_vislist
示例#5
0
def simulate_gaintable_from_voltage_pattern(vis,
                                            sc,
                                            vp,
                                            vis_slices=None,
                                            scale=1.0,
                                            order=3,
                                            use_radec=False,
                                            elevation_limit=15.0 * numpy.pi /
                                            180.0,
                                            **kwargs):
    """ Create gaintables from a list of components and voltagr patterns

    :param elevation_limit:
    :param use_radec:
    :param vis_slices:
    :param vis:
    :param sc: Sky components for which pierce points are needed
    :param vp: Voltage pattern in AZELGEO frame
    :param scale: Multiply the screen by this factor
    :param order: order of spline (default is 3)
    :return:
    """

    nant = vis.vis.shape[1]
    gaintables = [
        create_gaintable_from_blockvisibility(vis, **kwargs) for i in sc
    ]

    nchan, npol, ny, nx = vp.data.shape

    real_spline = [
        RectBivariateSpline(range(ny),
                            range(nx),
                            vp.data[0, pol, ...].real,
                            kx=order,
                            ky=order) for pol in range(npol)
    ]
    imag_spline = [
        RectBivariateSpline(range(ny),
                            range(nx),
                            vp.data[0, pol, ...].imag,
                            kx=order,
                            ky=order) for pol in range(npol)
    ]

    if not use_radec:
        assert isinstance(vis, BlockVisibility)
        assert vp.wcs.wcs.ctype[0] == 'AZELGEO long', vp.wcs.wcs.ctype[0]
        assert vp.wcs.wcs.ctype[1] == 'AZELGEO lati', vp.wcs.wcs.ctype[1]

        assert vis.configuration.mount[
            0] == 'azel', "Mount %s not supported yet" % vis.configuration.mount[
                0]

        number_bad = 0
        number_good = 0

        # For each hourangle, we need to calculate the location of a component
        # in AZELGEO. With that we can then look up the relevant gain from the
        # voltage pattern
        for iha, rows in enumerate(
                vis_timeslice_iter(vis, vis_slices=vis_slices)):
            v = create_visibility_from_rows(vis, rows)
            if v is not None:
                utc_time = Time([numpy.average(v.time) / 86400.0],
                                format='mjd',
                                scale='utc')
                azimuth_centre, elevation_centre = calculate_azel(
                    v.configuration.location, utc_time, vis.phasecentre)
                azimuth_centre = azimuth_centre[0].to('deg').value
                elevation_centre = elevation_centre[0].to('deg').value

                # Calculate the az el for this time
                wcs_azel = vp.wcs.sub(2).deepcopy()

                for icomp, comp in enumerate(sc):

                    if elevation_centre >= elevation_limit:

                        antgain = numpy.zeros([nant, npol], dtype='complex')
                        antwt = numpy.zeros([nant, npol])

                        # Calculate the azel of this component
                        azimuth_comp, elevation_comp = calculate_azel(
                            v.configuration.location, utc_time, comp.direction)
                        cosel = numpy.cos(elevation_comp[0]).value
                        azimuth_comp = azimuth_comp[0].to('deg').value
                        elevation_comp = elevation_comp[0].to('deg').value
                        if azimuth_comp - azimuth_centre > 180.0:
                            azimuth_centre += 360.0
                        elif azimuth_comp - azimuth_centre < -180.0:
                            azimuth_centre -= 360.0

                        try:
                            worldloc = [[
                                (azimuth_comp - azimuth_centre) * cosel,
                                elevation_comp - elevation_centre
                            ]]
                            # radius = numpy.sqrt(((azimuth_comp-azimuth_centre)*cosel)**2 +
                            #                     (elevation_comp-elevation_centre)**2)
                            pixloc = wcs_azel.wcs_world2pix(worldloc, 1)[0]
                            assert pixloc[0] > 2
                            assert pixloc[0] < nx - 3
                            assert pixloc[1] > 2
                            assert pixloc[1] < ny - 3
                            for pol in range(npol):
                                gain = real_spline[pol].ev(pixloc[1], pixloc[0]) \
                                       + 1j * imag_spline[pol].ev(pixloc[1], pixloc[0])
                                antgain[:, pol] = gain
                            for ant in range(nant):
                                ag = antgain[ant, :].reshape([2, 2])
                                ag = numpy.linalg.inv(ag)
                                antgain[ant, :] = ag.reshape([4])
                                number_good += 1
                        except (ValueError, AssertionError):
                            number_bad += 1
                            antgain[...] = 0.0
                            antwt[...] = 0.0

                        gaintables[icomp].gain[
                            iha, :, :, :] = antgain[:,
                                                    numpy.newaxis, :].reshape(
                                                        [nant, nchan, 2, 2])
                        gaintables[icomp].weight[
                            iha, :, :, :] = antwt[:, numpy.newaxis, :].reshape(
                                [nant, nchan, 2, 2])
                        gaintables[icomp].phasecentre = comp.direction
                    else:
                        gaintables[icomp].gain[...] = 1.0 + 0.0j
                        gaintables[icomp].weight[iha, :, :, :] = 0.0
                        gaintables[icomp].phasecentre = comp.direction
                        number_bad += nant

    else:
        assert isinstance(vis, BlockVisibility)
        assert vp.wcs.wcs.ctype[0] == 'RA---SIN', vp.wcs.wcs.ctype[0]
        assert vp.wcs.wcs.ctype[1] == 'DEC--SIN', vp.wcs.wcs.ctype[1]

        # The time in the Visibility is hour angle in seconds!
        number_bad = 0
        number_good = 0

        d2r = numpy.pi / 180.0
        ra_centre = vp.wcs.wcs.crval[0] * d2r
        dec_centre = vp.wcs.wcs.crval[1] * d2r

        r2d = 180.0 / numpy.pi
        s2r = numpy.pi / 43200.0
        # For each hourangle, we need to calculate the location of a component
        # in AZELGEO. With that we can then look up the relevant gain from the
        # voltage pattern
        for iha, rows in enumerate(
                vis_timeslice_iter(vis, vis_slices=vis_slices)):
            v = create_visibility_from_rows(vis, rows)
            ha = numpy.average(v.time)

            for icomp, comp in enumerate(sc):
                antgain = numpy.zeros([nant, npol], dtype='complex')
                antwt = numpy.zeros([nant, npol])
                # Calculate the location of the component in AZELGEO, then add the pointing offset
                # for each antenna
                ra_comp = comp.direction.ra.rad
                dec_comp = comp.direction.dec.rad
                for ant in range(nant):
                    wcs_azel = vp.wcs.deepcopy()
                    ra_pointing = ra_centre * r2d
                    dec_pointing = dec_centre * r2d

                    # We use WCS sensible coordinate handling by labelling the axes misleadingly
                    wcs_azel.wcs.crval[0] = ra_pointing
                    wcs_azel.wcs.crval[1] = dec_pointing
                    wcs_azel.wcs.ctype[0] = 'RA---SIN'
                    wcs_azel.wcs.ctype[1] = 'DEC--SIN'

                    for pol in range(npol):
                        worldloc = [
                            ra_comp * r2d, dec_comp * r2d, vp.wcs.wcs.crval[2],
                            vp.wcs.wcs.crval[3]
                        ]
                        try:
                            pixloc = wcs_azel.sub(2).wcs_world2pix(
                                [worldloc[:2]], 1)[0]
                            assert pixloc[0] > 2
                            assert pixloc[0] < nx - 3
                            assert pixloc[1] > 2
                            assert pixloc[1] < ny - 3
                            gain = real_spline[pol].ev(
                                pixloc[1], pixloc[0]
                            ) + 1j * imag_spline[pol].ev(pixloc[1], pixloc[0])
                            if numpy.abs(gain) > 0.0:
                                antgain[ant, pol] = 1.0 / (scale * gain)
                                antwt[ant, pol] = 1.0
                            else:
                                antgain[ant, pol] = 0.0
                                antwt[ant, pol] = 0.0
                            antwt[ant, pol] = 1.0
                            number_good += 1
                        except (ValueError, AssertionError):
                            number_bad += 1
                            antgain[ant, pol] = 1e15
                            antwt[ant, pol] = 0.0

                gaintables[icomp].gain[
                    iha, :, :, :] = antgain[:, numpy.newaxis, :].reshape(
                        [nant, nchan, 2, 2])
                gaintables[icomp].weight[
                    iha, :, :, :] = antwt[:, numpy.newaxis, :].reshape(
                        [nant, nchan, 2, 2])
                gaintables[icomp].phasecentre = comp.direction

    assert number_good > 0, "simulate_gaintable_from_voltage_pattern: No points inside the voltage pattern image"
    if number_bad > 0:
        log.warning(
            "simulate_gaintable_from_voltage_pattern: %d points are inside the voltage pattern image"
            % (number_good))
        log.warning(
            "simulate_gaintable_from_voltage_pattern: %d points are outside the voltage pattern image"
            % (number_bad))

    return gaintables
示例#6
0
def simulate_gaintable_from_zernikes(vis,
                                     sc,
                                     vp_list,
                                     vp_coeffs,
                                     vis_slices=None,
                                     order=3,
                                     use_radec=False,
                                     elevation_limit=15.0 * numpy.pi / 180.0,
                                     **kwargs):
    """ Create gaintables for a set of zernikes

    :param vis:
    :param sc: Sky components for which pierce points are needed
    :param vp: List of Voltage patterns in AZELGEO frame
    :param vp_coeffs: Fractional contribution [nants, nvp]
    :param order: order of spline (default is 3)
    :return:
    """

    ntimes, nant = vis.vis.shape[0:2]
    vp_coeffs = numpy.array(vp_coeffs)
    gaintables = [
        create_gaintable_from_blockvisibility(vis, **kwargs) for i in sc
    ]

    if not use_radec:
        assert isinstance(vis, BlockVisibility)
        assert vis.configuration.mount[
            0] == 'azel', "Mount %s not supported yet" % vis.configuration.mount[
                0]

        # The time in the Visibility is hour angle in seconds!
        number_bad = 0
        number_good = 0

        # Cache the splines, one per voltage pattern
        real_splines = list()
        imag_splines = list()
        for ivp, vp in enumerate(vp_list):
            assert vp.wcs.wcs.ctype[0] == 'AZELGEO long', vp.wcs.wcs.ctype[0]
            assert vp.wcs.wcs.ctype[1] == 'AZELGEO lati', vp.wcs.wcs.ctype[1]

            nchan, npol, ny, nx = vp.data.shape
            real_splines.append(
                RectBivariateSpline(range(ny),
                                    range(nx),
                                    vp.data[0, 0, ...].real,
                                    kx=order,
                                    ky=order))
            imag_splines.append(
                RectBivariateSpline(range(ny),
                                    range(nx),
                                    vp.data[0, 0, ...].imag,
                                    kx=order,
                                    ky=order))

        latitude = vis.configuration.location.lat.rad

        r2d = 180.0 / numpy.pi
        s2r = numpy.pi / 43200.0
        # For each hourangle, we need to calculate the location of a component
        # in AZELGEO. With that we can then look up the relevant gain from the
        # voltage pattern
        for iha, rows in enumerate(
                vis_timeslice_iter(vis, vis_slices=vis_slices)):
            v = create_visibility_from_rows(vis, rows)
            ha = numpy.average(
                calculate_blockvisibility_hourangles(v).to('rad').value)

            # Calculate the az el for this hourangle and the phasecentre declination
            utc_time = Time([numpy.average(v.time) / 86400.0],
                            format='mjd',
                            scale='utc')
            azimuth_centre, elevation_centre = calculate_azel(
                v.configuration.location, utc_time, vis.phasecentre)
            azimuth_centre = azimuth_centre[0].to('deg').value
            elevation_centre = elevation_centre[0].to('deg').value

            for icomp, comp in enumerate(sc):

                if elevation_centre >= elevation_limit:

                    antgain = numpy.zeros([nant], dtype='complex')
                    # Calculate the location of the component in AZELGEO, then add the pointing offset
                    # for each antenna
                    hacomp = comp.direction.ra.rad - vis.phasecentre.ra.rad + ha
                    deccomp = comp.direction.dec.rad
                    azimuth_comp, elevation_comp = hadec_to_azel(
                        hacomp, deccomp, latitude)

                    for ant in range(nant):
                        for ivp, vp in enumerate(vp_list):
                            nchan, npol, ny, nx = vp.data.shape
                            wcs_azel = vp.wcs.deepcopy()

                            # We use WCS sensible coordinate handling by labelling the axes misleadingly
                            wcs_azel.wcs.crval[0] = azimuth_centre
                            wcs_azel.wcs.crval[1] = elevation_centre
                            wcs_azel.wcs.ctype[0] = 'RA---SIN'
                            wcs_azel.wcs.ctype[1] = 'DEC--SIN'

                            worldloc = [
                                azimuth_comp * r2d, elevation_comp * r2d,
                                vp.wcs.wcs.crval[2], vp.wcs.wcs.crval[3]
                            ]
                            try:
                                pixloc = wcs_azel.sub(2).wcs_world2pix(
                                    [worldloc[:2]], 1)[0]
                                assert pixloc[0] > 2
                                assert pixloc[0] < nx - 3
                                assert pixloc[1] > 2
                                assert pixloc[1] < ny - 3
                                gain = real_splines[ivp].ev(pixloc[1], pixloc[0]) \
                                       + 1j * imag_splines[ivp](pixloc[1], pixloc[0])
                                antgain[ant] += vp_coeffs[ant, ivp] * gain
                                number_good += 1
                            except (ValueError, AssertionError):
                                number_bad += 1
                                antgain[ant] = 1.0

                        antgain[ant] = 1.0 / antgain[ant]

                    gaintables[icomp].gain[
                        iha, :, :, :] = antgain[:, numpy.newaxis,
                                                numpy.newaxis, numpy.newaxis]
                    gaintables[icomp].phasecentre = comp.direction
            else:
                gaintables[icomp].gain[...] = 1.0 + 0.0j
                gaintables[icomp].phasecentre = comp.direction
                number_bad += nant

    else:
        assert isinstance(vis, BlockVisibility)
        number_bad = 0
        number_good = 0

        # Cache the splines, one per voltage pattern
        real_splines = list()
        imag_splines = list()
        for ivp, vp in enumerate(vp_list):
            nchan, npol, ny, nx = vp.data.shape
            real_splines.append(
                RectBivariateSpline(range(ny),
                                    range(nx),
                                    vp.data[0, 0, ...].real,
                                    kx=order,
                                    ky=order))
            imag_splines.append(
                RectBivariateSpline(range(ny),
                                    range(nx),
                                    vp.data[0, 0, ...].imag,
                                    kx=order,
                                    ky=order))

        for iha, rows in enumerate(
                vis_timeslice_iter(vis, vis_slices=vis_slices)):

            # The time in the Visibility is hour angle in seconds!
            r2d = 180.0 / numpy.pi
            # For each hourangle, we need to calculate the location of a component
            # in AZELGEO. With that we can then look up the relevant gain from the
            # voltage pattern
            v = create_visibility_from_rows(vis, rows)
            ha = numpy.average(calculate_blockvisibility_hourangles(v))

            for icomp, comp in enumerate(sc):
                antgain = numpy.zeros([nant], dtype='complex')
                antwt = numpy.zeros([nant])
                ra_comp = comp.direction.ra.rad
                dec_comp = comp.direction.dec.rad
                for ant in range(nant):
                    for ivp, vp in enumerate(vp_list):

                        assert vp.wcs.wcs.ctype[
                            0] == 'RA---SIN', vp.wcs.wcs.ctype[0]
                        assert vp.wcs.wcs.ctype[
                            1] == 'DEC--SIN', vp.wcs.wcs.ctype[1]

                        worldloc = [
                            ra_comp * r2d, dec_comp * r2d, vp.wcs.wcs.crval[2],
                            vp.wcs.wcs.crval[3]
                        ]
                        nchan, npol, ny, nx = vp.data.shape

                        try:
                            pixloc = vp.wcs.sub(2).wcs_world2pix(
                                [worldloc[:2]], 1)[0]
                            assert pixloc[0] > 2
                            assert pixloc[0] < nx - 3
                            assert pixloc[1] > 2
                            assert pixloc[1] < ny - 3
                            gain = real_splines[ivp].ev(pixloc[1], pixloc[0]) \
                                   + 1j * imag_splines[ivp](pixloc[1], pixloc[0])
                            antgain[ant] += vp_coeffs[ant, ivp] * gain
                            antwt[ant] = 1.0
                            number_good += 1
                        except (ValueError, AssertionError):
                            number_bad += 1
                            antgain[ant] = 1e15
                            antwt[ant] = 0.0

                        antgain[ant] = 1.0 / antgain[ant]

                    gaintables[icomp].gain[
                        iha, :, :, :] = antgain[:, numpy.newaxis,
                                                numpy.newaxis, numpy.newaxis]
                    gaintables[icomp].weight[
                        iha, :, :, :] = antwt[:, numpy.newaxis, numpy.newaxis,
                                              numpy.newaxis]
                    gaintables[icomp].phasecentre = comp.direction

    if number_bad > 0:
        log.warning(
            "simulate_gaintable_from_zernikes: %d points are inside the voltage pattern image"
            % (number_good))
        log.warning(
            "simulate_gaintable_from_zernikes: %d points are outside the voltage pattern image"
            % (number_bad))

    return gaintables
示例#7
0
    def test_apply_voltage_pattern_image_pointsource(self):
        self.createVis(rmax=1e3)
        telescope = 'MID_FEKO_B2'
        vpol = PolarisationFrame("linear")
        self.times = numpy.linspace(-4, +4, 8) * numpy.pi / 12.0
        bvis = create_blockvisibility(self.config,
                                      self.times,
                                      self.frequency,
                                      channel_bandwidth=self.channel_bandwidth,
                                      phasecentre=self.phasecentre,
                                      weight=1.0,
                                      polarisation_frame=vpol,
                                      zerow=True)
        cellsize = advise_wide_field(bvis)['cellsize']

        pbmodel = create_image_from_visibility(
            bvis,
            cellsize=self.cellsize,
            npixel=self.npixel,
            override_cellsize=False,
            polarisation_frame=PolarisationFrame("stokesIQUV"))
        vpbeam = create_vp(pbmodel, telescope=telescope, use_local=False)
        vpbeam.wcs.wcs.ctype[0] = 'RA---SIN'
        vpbeam.wcs.wcs.ctype[1] = 'DEC--SIN'
        vpbeam.wcs.wcs.crval[0] = pbmodel.wcs.wcs.crval[0]
        vpbeam.wcs.wcs.crval[1] = pbmodel.wcs.wcs.crval[1]

        s3_components = create_test_skycomponents_from_s3(
            flux_limit=0.1,
            phasecentre=self.phasecentre,
            frequency=self.frequency,
            polarisation_frame=PolarisationFrame('stokesI'),
            radius=1.5 * numpy.pi / 180.0)

        for comp in s3_components:
            comp.polarisation_frame = PolarisationFrame('stokesIQUV')
            comp.flux = numpy.array([[comp.flux[0, 0], 0.0, 0.0, 0.0]])

        s3_components = filter_skycomponents_by_flux(s3_components, 0.0, 10.0)

        from rascil.processing_components.image import show_image
        import matplotlib.pyplot as plt
        plt.clf()
        show_image(vpbeam, components=s3_components)
        plt.show(block=False)

        vpcomp = apply_voltage_pattern_to_skycomponent(s3_components, vpbeam)
        bvis.data['vis'][...] = 0.0 + 0.0j
        bvis = dft_skycomponent_visibility(bvis, vpcomp)

        rec_comp = idft_visibility_skycomponent(bvis, vpcomp)[0]

        stokes_comp = list()
        for comp in rec_comp:
            stokes_comp.append(
                convert_pol_frame(comp.flux[0], PolarisationFrame("linear"),
                                  PolarisationFrame("stokesIQUV")))

        stokesI = numpy.abs(
            numpy.array([comp_flux[0] for comp_flux in stokes_comp]).real)
        stokesQ = numpy.abs(
            numpy.array([comp_flux[1] for comp_flux in stokes_comp]).real)
        stokesU = numpy.abs(
            numpy.array([comp_flux[2] for comp_flux in stokes_comp]).real)
        stokesV = numpy.abs(
            numpy.array([comp_flux[3] for comp_flux in stokes_comp]).real)
        plt.clf()
        plt.loglog(stokesI, stokesQ, '.', label='Q')
        plt.loglog(stokesI, stokesU, '.', label='U')
        plt.loglog(stokesI, stokesV, '.', label='V')
        plt.xlabel("Stokes Flux I (Jy)")
        plt.ylabel("Flux (Jy)")
        plt.legend()
        plt.savefig('%s/test_primary_beams_pol_rsexecute_stokes_errors.png' %
                    self.dir)
        plt.show(block=False)

        split_times = False
        if split_times:
            bvis_list = list()
            for rows in vis_timeslice_iter(bvis, vis_slices=8):
                bvis_list.append(create_visibility_from_rows(bvis, rows))
        else:
            bvis_list = [bvis]

        bvis_list = rsexecute.scatter(bvis_list)

        model_list = \
            [rsexecute.execute(create_image_from_visibility, nout=1)(bv, cellsize=cellsize, npixel=4096,
                                                                     phasecentre=self.phasecentre,
                                                                     override_cellsize=False,
                                                                     polarisation_frame=PolarisationFrame("stokesIQUV"))
             for bv in bvis_list]

        model_list = rsexecute.persist(model_list)
        bvis_list = weight_list_rsexecute_workflow(bvis_list, model_list)

        continuum_imaging_list = \
            continuum_imaging_list_rsexecute_workflow(bvis_list, model_list,
                                                      context='2d',
                                                      algorithm='hogbom',
                                                      facets=1,
                                                      niter=1000,
                                                      fractional_threshold=0.1,
                                                      threshold=1e-4,
                                                      nmajor=5, gain=0.1,
                                                      deconvolve_facets=4,
                                                      deconvolve_overlap=32,
                                                      deconvolve_taper='tukey',
                                                      psf_support=64,
                                                      restore_facets=4, psfwidth=1.0)
        clean, residual, restored = rsexecute.compute(continuum_imaging_list,
                                                      sync=True)
        centre = 0
        if self.persist:
            export_image_to_fits(
                clean[centre],
                '%s/test_primary_beams_pol_rsexecute_clean.fits' % self.dir)
            export_image_to_fits(
                residual[centre][0],
                '%s/test_primary_beams_pol_rsexecute_residual.fits' % self.dir)
            export_image_to_fits(
                restored[centre],
                '%s/test_primary_beams_pol_rsexecute_restored.fits' % self.dir)

        plt.clf()
        show_image(restored[centre])
        plt.show(block=False)

        qa = qa_image(restored[centre])
        assert numpy.abs(qa.data['max'] - 0.9953017707113947) < 1.0e-7, str(qa)
        assert numpy.abs(qa.data['min'] +
                         0.0036396480874570846) < 1.0e-7, str(qa)