[2] | 1 | //#---------------------------------------------------------------------------
|
---|
| 2 | //# SDMath.cc: A collection of single dish mathematical operations
|
---|
| 3 | //#---------------------------------------------------------------------------
|
---|
| 4 | //# Copyright (C) 2004
|
---|
[125] | 5 | //# ATNF
|
---|
[2] | 6 | //#
|
---|
| 7 | //# This program is free software; you can redistribute it and/or modify it
|
---|
| 8 | //# under the terms of the GNU General Public License as published by the Free
|
---|
| 9 | //# Software Foundation; either version 2 of the License, or (at your option)
|
---|
| 10 | //# any later version.
|
---|
| 11 | //#
|
---|
| 12 | //# This program is distributed in the hope that it will be useful, but
|
---|
| 13 | //# WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 14 | //# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
|
---|
| 15 | //# Public License for more details.
|
---|
| 16 | //#
|
---|
| 17 | //# You should have received a copy of the GNU General Public License along
|
---|
| 18 | //# with this program; if not, write to the Free Software Foundation, Inc.,
|
---|
| 19 | //# 675 Massachusetts Ave, Cambridge, MA 02139, USA.
|
---|
| 20 | //#
|
---|
| 21 | //# Correspondence concerning this software should be addressed as follows:
|
---|
| 22 | //# Internet email: Malte.Marquarding@csiro.au
|
---|
| 23 | //# Postal address: Malte Marquarding,
|
---|
| 24 | //# Australia Telescope National Facility,
|
---|
| 25 | //# P.O. Box 76,
|
---|
| 26 | //# Epping, NSW, 2121,
|
---|
| 27 | //# AUSTRALIA
|
---|
| 28 | //#
|
---|
| 29 | //# $Id:
|
---|
| 30 | //#---------------------------------------------------------------------------
|
---|
[38] | 31 | #include <vector>
|
---|
| 32 |
|
---|
[81] | 33 | #include <casa/aips.h>
|
---|
| 34 | #include <casa/BasicSL/String.h>
|
---|
| 35 | #include <casa/Arrays/IPosition.h>
|
---|
| 36 | #include <casa/Arrays/Array.h>
|
---|
[130] | 37 | #include <casa/Arrays/ArrayIter.h>
|
---|
| 38 | #include <casa/Arrays/VectorIter.h>
|
---|
[81] | 39 | #include <casa/Arrays/ArrayMath.h>
|
---|
| 40 | #include <casa/Arrays/ArrayLogical.h>
|
---|
| 41 | #include <casa/Arrays/MaskedArray.h>
|
---|
| 42 | #include <casa/Arrays/MaskArrMath.h>
|
---|
| 43 | #include <casa/Arrays/MaskArrLogi.h>
|
---|
[130] | 44 | #include <casa/Exceptions.h>
|
---|
[2] | 45 |
|
---|
[81] | 46 | #include <tables/Tables/Table.h>
|
---|
| 47 | #include <tables/Tables/ScalarColumn.h>
|
---|
| 48 | #include <tables/Tables/ArrayColumn.h>
|
---|
[2] | 49 |
|
---|
[130] | 50 | #include <lattices/Lattices/LatticeUtilities.h>
|
---|
| 51 | #include <lattices/Lattices/RebinLattice.h>
|
---|
[81] | 52 | #include <coordinates/Coordinates/SpectralCoordinate.h>
|
---|
[130] | 53 | #include <coordinates/Coordinates/CoordinateSystem.h>
|
---|
| 54 | #include <coordinates/Coordinates/CoordinateUtil.h>
|
---|
[38] | 55 |
|
---|
| 56 | #include "MathUtils.h"
|
---|
[2] | 57 | #include "SDContainer.h"
|
---|
| 58 | #include "SDMemTable.h"
|
---|
| 59 |
|
---|
| 60 | #include "SDMath.h"
|
---|
| 61 |
|
---|
[125] | 62 | using namespace casa;
|
---|
[83] | 63 | using namespace asap;
|
---|
| 64 | //using namespace asap::SDMath;
|
---|
[2] | 65 |
|
---|
[144] | 66 | CountedPtr<SDMemTable> SDMath::average (const Block<CountedPtr<SDMemTable> >& in,
|
---|
| 67 | const Vector<Bool>& mask, bool scanAv,
|
---|
| 68 | const std::string& weightStr)
|
---|
[130] | 69 | //
|
---|
[144] | 70 | // Weighted averaging of spectra from one or more Tables.
|
---|
[130] | 71 | //
|
---|
| 72 | {
|
---|
[144] | 73 | weightType wtType = NONE;
|
---|
| 74 | String tStr(weightStr);
|
---|
| 75 | tStr.upcase();
|
---|
| 76 | if (tStr.contains(String("NONE"))) {
|
---|
| 77 | wtType = NONE;
|
---|
| 78 | } else if (tStr.contains(String("VAR"))) {
|
---|
| 79 | wtType = VAR;
|
---|
| 80 | } else if (tStr.contains(String("TSYS"))) {
|
---|
| 81 | wtType = TSYS;
|
---|
| 82 | throw (AipsError("T_sys weighting not yet implemented"));
|
---|
| 83 | } else {
|
---|
| 84 | throw (AipsError("Unrecognized weighting type"));
|
---|
| 85 | }
|
---|
[2] | 86 |
|
---|
[144] | 87 | // Create output Table by cloning from the first table
|
---|
[2] | 88 |
|
---|
[144] | 89 | SDMemTable* pTabOut = new SDMemTable(*in[0],True);
|
---|
[130] | 90 |
|
---|
[144] | 91 | // Setup
|
---|
[130] | 92 |
|
---|
[144] | 93 | const uInt axis = 3; // Spectral axis
|
---|
| 94 | IPosition shp = in[0]->rowAsMaskedArray(0).shape(); // Must not change
|
---|
| 95 | Array<Float> arr(shp);
|
---|
| 96 | Array<Bool> barr(shp);
|
---|
| 97 | const Bool useMask = (mask.nelements() == shp(axis));
|
---|
[130] | 98 |
|
---|
[144] | 99 | // Columns from Tables
|
---|
[130] | 100 |
|
---|
[144] | 101 | ROArrayColumn<Float> tSysCol;
|
---|
| 102 | ROScalarColumn<Double> mjdCol;
|
---|
| 103 | ROScalarColumn<String> srcNameCol;
|
---|
| 104 | ROScalarColumn<Double> intCol;
|
---|
| 105 | ROArrayColumn<uInt> fqIDCol;
|
---|
[130] | 106 |
|
---|
[144] | 107 | // Create accumulation MaskedArray. We accumulate for each channel,if,pol,beam
|
---|
| 108 | // Note that the mask of the accumulation array will ALWAYS remain ALL True.
|
---|
| 109 | // The MA is only used so that when data which is masked Bad is added to it,
|
---|
| 110 | // that data does not contribute.
|
---|
| 111 |
|
---|
| 112 | Array<Float> zero(shp);
|
---|
| 113 | zero=0.0;
|
---|
| 114 | Array<Bool> good(shp);
|
---|
| 115 | good = True;
|
---|
| 116 | MaskedArray<Float> sum(zero,good);
|
---|
| 117 |
|
---|
| 118 | // Counter arrays
|
---|
| 119 |
|
---|
| 120 | Array<Float> nPts(shp); // Number of points
|
---|
| 121 | nPts = 0.0;
|
---|
| 122 | Array<Float> nInc(shp); // Increment
|
---|
| 123 | nInc = 1.0;
|
---|
| 124 |
|
---|
| 125 | // Create accumulation Array for variance. We accumulate for
|
---|
| 126 | // each if,pol,beam, but average over channel. So we need
|
---|
| 127 | // a shape with one less axis dropping channels.
|
---|
| 128 |
|
---|
| 129 | const uInt nAxesSub = shp.nelements() - 1;
|
---|
| 130 | IPosition shp2(nAxesSub);
|
---|
| 131 | for (uInt i=0,j=0; i<(nAxesSub+1); i++) {
|
---|
| 132 | if (i!=axis) {
|
---|
| 133 | shp2(j) = shp(i);
|
---|
| 134 | j++;
|
---|
| 135 | }
|
---|
[2] | 136 | }
|
---|
[144] | 137 | Array<Float> sumSq(shp2);
|
---|
| 138 | sumSq = 0.0;
|
---|
| 139 | IPosition pos2(nAxesSub,0); // For indexing
|
---|
[130] | 140 |
|
---|
[144] | 141 | // Time-related accumulators
|
---|
[130] | 142 |
|
---|
[144] | 143 | Double time;
|
---|
| 144 | Double timeSum = 0.0;
|
---|
| 145 | Double intSum = 0.0;
|
---|
| 146 | Double interval = 0.0;
|
---|
[130] | 147 |
|
---|
[144] | 148 | // To get the right shape for the Tsys accumulator we need to
|
---|
| 149 | // access a column from the first table. The shape of this
|
---|
| 150 | // array must not change
|
---|
[130] | 151 |
|
---|
[144] | 152 | Array<Float> tSysSum;
|
---|
| 153 | {
|
---|
| 154 | const Table& tabIn = in[0]->table();
|
---|
| 155 | tSysCol.attach(tabIn,"TSYS");
|
---|
| 156 | tSysSum.resize(tSysCol.shape(0));
|
---|
| 157 | }
|
---|
| 158 | tSysSum =0.0;
|
---|
| 159 | Array<Float> tSys;
|
---|
| 160 |
|
---|
| 161 | // Scan and row tracking
|
---|
| 162 |
|
---|
| 163 | Int oldScanID = 0;
|
---|
| 164 | Int outScanID = 0;
|
---|
| 165 | Int scanID = 0;
|
---|
| 166 | Int rowStart = 0;
|
---|
| 167 | Int nAccum = 0;
|
---|
| 168 | Int tableStart = 0;
|
---|
| 169 |
|
---|
| 170 | // Source and FreqID
|
---|
| 171 |
|
---|
| 172 | String sourceName, oldSourceName, sourceNameStart;
|
---|
| 173 | Vector<uInt> freqID, freqIDStart, oldFreqID;
|
---|
| 174 |
|
---|
| 175 | // Loop over tables
|
---|
| 176 |
|
---|
| 177 | Float fac = 1.0;
|
---|
| 178 | const uInt nTables = in.nelements();
|
---|
| 179 | for (uInt iTab=0; iTab<nTables; iTab++) {
|
---|
| 180 |
|
---|
| 181 | // Attach columns to Table
|
---|
| 182 |
|
---|
| 183 | const Table& tabIn = in[iTab]->table();
|
---|
| 184 | tSysCol.attach(tabIn, "TSYS");
|
---|
| 185 | mjdCol.attach(tabIn, "TIME");
|
---|
| 186 | srcNameCol.attach(tabIn, "SRCNAME");
|
---|
| 187 | intCol.attach(tabIn, "INTERVAL");
|
---|
| 188 | fqIDCol.attach(tabIn, "FREQID");
|
---|
| 189 |
|
---|
| 190 | // Loop over rows in Table
|
---|
| 191 |
|
---|
| 192 | const uInt nRows = in[iTab]->nRow();
|
---|
| 193 | for (uInt iRow=0; iRow<nRows; iRow++) {
|
---|
| 194 |
|
---|
| 195 | // Check conformance
|
---|
| 196 |
|
---|
| 197 | IPosition shp2 = in[iTab]->rowAsMaskedArray(iRow).shape();
|
---|
| 198 | if (!shp.isEqual(shp2)) {
|
---|
| 199 | throw (AipsError("Shapes for all rows must be the same"));
|
---|
| 200 | }
|
---|
| 201 |
|
---|
| 202 | // If we are not doing scan averages, make checks for source and
|
---|
| 203 | // frequency setup and warn if averaging across them
|
---|
| 204 |
|
---|
| 205 | // Get copy of Scan Container for this row
|
---|
| 206 |
|
---|
| 207 | SDContainer sc = in[iTab]->getSDContainer(iRow);
|
---|
| 208 | scanID = sc.scanid;
|
---|
| 209 |
|
---|
| 210 | // Get quantities from columns
|
---|
| 211 |
|
---|
| 212 | srcNameCol.getScalar(iRow, sourceName);
|
---|
| 213 | mjdCol.get(iRow, time);
|
---|
| 214 | tSysCol.get(iRow, tSys);
|
---|
| 215 | intCol.get(iRow, interval);
|
---|
| 216 | fqIDCol.get(iRow, freqID);
|
---|
| 217 |
|
---|
| 218 | // Initialize first source and freqID
|
---|
| 219 |
|
---|
| 220 | if (iRow==0 && iTab==0) {
|
---|
| 221 | sourceNameStart = sourceName;
|
---|
| 222 | freqIDStart = freqID;
|
---|
| 223 | }
|
---|
| 224 |
|
---|
| 225 | // If we are doing scan averages, see if we are at the end of an
|
---|
| 226 | // accumulation period (scan). We must check soutce names too,
|
---|
| 227 | // since we might have two tables with one scan each but different
|
---|
| 228 | // source names; we shouldn't average different sources together
|
---|
| 229 |
|
---|
| 230 | if (scanAv && ( (scanID != oldScanID) ||
|
---|
| 231 | (iRow==0 && iTab>0 && sourceName!=oldSourceName))) {
|
---|
| 232 |
|
---|
| 233 | // Normalize data in 'sum' accumulation array according to weighting scheme
|
---|
| 234 |
|
---|
| 235 | normalize (sum, sumSq, nPts, wtType, axis, nAxesSub);
|
---|
| 236 |
|
---|
| 237 | // Fill scan container. The source and freqID come from the
|
---|
| 238 | // first row of the first table that went into this average (
|
---|
| 239 | // should be the same for all rows in the scan average)
|
---|
| 240 |
|
---|
| 241 | Float nR(nAccum);
|
---|
| 242 | fillSDC (sc, sum.getMask(), sum.getArray(), tSysSum/nR, outScanID,
|
---|
| 243 | timeSum/nR, intSum, sourceNameStart, freqIDStart);
|
---|
| 244 |
|
---|
| 245 | // Write container out to Table
|
---|
| 246 |
|
---|
| 247 | pTabOut->putSDContainer(sc);
|
---|
| 248 |
|
---|
| 249 | // Reset accumulators
|
---|
| 250 |
|
---|
| 251 | sum = 0.0;
|
---|
| 252 | sumSq = 0.0;
|
---|
| 253 | nAccum = 0;
|
---|
| 254 | //
|
---|
| 255 | tSysSum =0.0;
|
---|
| 256 | timeSum = 0.0;
|
---|
| 257 | intSum = 0.0;
|
---|
| 258 |
|
---|
| 259 | // Increment
|
---|
| 260 |
|
---|
| 261 | rowStart = iRow; // First row for next accumulation
|
---|
| 262 | tableStart = iTab; // First table for next accumulation
|
---|
| 263 | sourceNameStart = sourceName; // First source name for next accumulation
|
---|
| 264 | freqIDStart = freqID; // First FreqID for next accumulation
|
---|
| 265 | //
|
---|
| 266 | oldScanID = scanID;
|
---|
| 267 | outScanID += 1; // Scan ID for next accumulation period
|
---|
| 268 | }
|
---|
| 269 |
|
---|
| 270 | // Accumulation step. First get data and deconstruct
|
---|
| 271 |
|
---|
| 272 | MaskedArray<Float> dataIn(in[iTab]->rowAsMaskedArray(iRow));
|
---|
| 273 | Array<Float>& valuesIn = dataIn.getRWArray(); // writable reference
|
---|
| 274 | const Array<Bool>& maskIn = dataIn.getMask(); // RO reference
|
---|
| 275 | //
|
---|
| 276 | if (wtType==NONE) {
|
---|
| 277 | const MaskedArray<Float> n(nInc,dataIn.getMask());
|
---|
| 278 | nPts += n; // Only accumulates where mask==T
|
---|
| 279 | } else if (wtType==VAR) {
|
---|
| 280 |
|
---|
| 281 | // We are going to average the data, weighted by the noise for each pol, beam and IF.
|
---|
| 282 | // So therefore we need to iterate through by spectrum (axis 3)
|
---|
| 283 |
|
---|
| 284 | VectorIterator<Float> itData(valuesIn, axis);
|
---|
| 285 | ReadOnlyVectorIterator<Bool> itMask(maskIn, axis);
|
---|
| 286 | while (!itData.pastEnd()) {
|
---|
| 287 |
|
---|
| 288 | // Make MaskedArray of Vector, optionally apply OTF mask, and find scaling factor
|
---|
| 289 |
|
---|
| 290 | if (useMask) {
|
---|
| 291 | MaskedArray<Float> tmp(itData.vector(),mask&&itMask.vector());
|
---|
| 292 | fac = 1.0/variance(tmp);
|
---|
| 293 | } else {
|
---|
| 294 | MaskedArray<Float> tmp(itData.vector(),itMask.vector());
|
---|
| 295 | fac = 1.0/variance(tmp);
|
---|
| 296 | }
|
---|
| 297 |
|
---|
| 298 | // Scale data
|
---|
| 299 |
|
---|
| 300 | itData.vector() *= fac; // Writes back into 'dataIn'
|
---|
| 301 | //
|
---|
| 302 | // Accumulate variance per if/pol/beam averaged over spectrum
|
---|
| 303 | // This method to get pos2 from itData.pos() is only valid
|
---|
| 304 | // because the spectral axis is the last one (so we can just
|
---|
| 305 | // copy the first nAXesSub positions out)
|
---|
| 306 |
|
---|
| 307 | pos2 = itData.pos().getFirst(nAxesSub);
|
---|
| 308 | sumSq(pos2) += fac;
|
---|
| 309 | //
|
---|
| 310 | itData.next();
|
---|
| 311 | itMask.next();
|
---|
| 312 | }
|
---|
| 313 | } else if (wtType==TSYS) {
|
---|
| 314 | }
|
---|
| 315 |
|
---|
| 316 | // Accumulate sum of (possibly scaled) data
|
---|
| 317 |
|
---|
| 318 | sum += dataIn;
|
---|
| 319 |
|
---|
| 320 | // Accumulate Tsys, time, and interval
|
---|
| 321 |
|
---|
| 322 | tSysSum += tSys;
|
---|
| 323 | timeSum += time;
|
---|
| 324 | intSum += interval;
|
---|
| 325 |
|
---|
| 326 | // Number of rows in accumulation
|
---|
| 327 |
|
---|
| 328 | nAccum += 1;
|
---|
| 329 | oldSourceName = sourceName;
|
---|
| 330 | oldFreqID = freqID;
|
---|
| 331 | }
|
---|
| 332 | }
|
---|
| 333 |
|
---|
| 334 | // OK at this point we have accumulation data which is either
|
---|
| 335 | // - accumulated from all tables into one row
|
---|
| 336 | // or
|
---|
| 337 | // - accumulated from the last scan average
|
---|
| 338 | //
|
---|
| 339 | // Normalize data in 'sum' accumulation array according to weighting scheme
|
---|
| 340 |
|
---|
| 341 | normalize (sum, sumSq, nPts, wtType, axis, nAxesSub);
|
---|
| 342 |
|
---|
| 343 | // Create and fill container. The container we clone will be from
|
---|
| 344 | // the last Table and the first row that went into the current
|
---|
| 345 | // accumulation. It probably doesn't matter that much really...
|
---|
| 346 |
|
---|
| 347 | Float nR(nAccum);
|
---|
| 348 | SDContainer sc = in[tableStart]->getSDContainer(rowStart);
|
---|
| 349 | fillSDC (sc, sum.getMask(), sum.getArray(), tSysSum/nR, outScanID,
|
---|
| 350 | timeSum/nR, intSum, sourceNameStart, freqIDStart);
|
---|
| 351 | //
|
---|
| 352 | pTabOut->putSDContainer(sc);
|
---|
| 353 | /*
|
---|
| 354 | cout << endl;
|
---|
| 355 | cout << "Last accumulation for output scan ID " << outScanID << endl;
|
---|
| 356 | cout << " The first row in this accumulation is " << rowStart << endl;
|
---|
| 357 | cout << " The number of rows accumulated is " << nAccum << endl;
|
---|
| 358 | cout << " The first table in this accumulation is " << tableStart << endl;
|
---|
| 359 | cout << " The first source in this accumulation is " << sourceNameStart << endl;
|
---|
| 360 | cout << " The first freqID in this accumulation is " << freqIDStart << endl;
|
---|
| 361 | cout << " Average time stamp = " << timeSum/nR << endl;
|
---|
| 362 | cout << " Integrated time = " << intSum << endl;
|
---|
| 363 | */
|
---|
| 364 | return CountedPtr<SDMemTable>(pTabOut);
|
---|
[2] | 365 | }
|
---|
[9] | 366 |
|
---|
[144] | 367 |
|
---|
| 368 |
|
---|
[85] | 369 | CountedPtr<SDMemTable>
|
---|
| 370 | SDMath::quotient(const CountedPtr<SDMemTable>& on,
|
---|
[130] | 371 | const CountedPtr<SDMemTable>& off)
|
---|
| 372 | //
|
---|
| 373 | // Compute quotient spectrum
|
---|
| 374 | //
|
---|
| 375 | {
|
---|
| 376 | const uInt nRows = on->nRow();
|
---|
| 377 | if (off->nRow() != nRows) {
|
---|
| 378 | throw (AipsError("Input Scan Tables must have the same number of rows"));
|
---|
| 379 | }
|
---|
[85] | 380 |
|
---|
[130] | 381 | // Input Tables and columns
|
---|
| 382 |
|
---|
[9] | 383 | Table ton = on->table();
|
---|
| 384 | Table toff = off->table();
|
---|
[85] | 385 | ROArrayColumn<Float> tsys(toff, "TSYS");
|
---|
[9] | 386 | ROScalarColumn<Double> mjd(ton, "TIME");
|
---|
[15] | 387 | ROScalarColumn<Double> integr(ton, "INTERVAL");
|
---|
[9] | 388 | ROScalarColumn<String> srcn(ton, "SRCNAME");
|
---|
[38] | 389 | ROArrayColumn<uInt> freqidc(ton, "FREQID");
|
---|
| 390 |
|
---|
[130] | 391 | // Output Table cloned from input
|
---|
[85] | 392 |
|
---|
[15] | 393 | SDMemTable* sdmt = new SDMemTable(*on, True);
|
---|
[130] | 394 |
|
---|
| 395 | // Loop over rows
|
---|
| 396 |
|
---|
| 397 | for (uInt i=0; i<nRows; i++) {
|
---|
| 398 | MaskedArray<Float> mon(on->rowAsMaskedArray(i));
|
---|
| 399 | MaskedArray<Float> moff(off->rowAsMaskedArray(i));
|
---|
| 400 | IPosition ipon = mon.shape();
|
---|
| 401 | IPosition ipoff = moff.shape();
|
---|
| 402 | //
|
---|
| 403 | Array<Float> tsarr;
|
---|
| 404 | tsys.get(i, tsarr);
|
---|
| 405 | if (ipon != ipoff && ipon != tsarr.shape()) {
|
---|
| 406 | throw(AipsError("on/off not conformant"));
|
---|
| 407 | }
|
---|
| 408 |
|
---|
| 409 | // Compute quotient
|
---|
| 410 |
|
---|
| 411 | MaskedArray<Float> tmp = (mon-moff);
|
---|
| 412 | Array<Float> out(tmp.getArray());
|
---|
| 413 | out /= moff;
|
---|
| 414 | out *= tsarr;
|
---|
| 415 | Array<Bool> outflagsb = !(mon.getMask() && moff.getMask());
|
---|
| 416 | Array<uChar> outflags(outflagsb.shape());
|
---|
| 417 | convertArray(outflags,outflagsb);
|
---|
| 418 |
|
---|
| 419 | // Fill container for this row
|
---|
| 420 |
|
---|
| 421 | SDContainer sc = on->getSDContainer();
|
---|
| 422 | sc.putTsys(tsarr);
|
---|
| 423 | sc.scanid = 0;
|
---|
| 424 | sc.putSpectrum(out);
|
---|
| 425 | sc.putFlags(outflags);
|
---|
| 426 |
|
---|
| 427 | // Put new row in output Table
|
---|
| 428 |
|
---|
| 429 | sdmt->putSDContainer(sc);
|
---|
| 430 | }
|
---|
| 431 | //
|
---|
[9] | 432 | return CountedPtr<SDMemTable>(sdmt);
|
---|
| 433 | }
|
---|
[48] | 434 |
|
---|
[139] | 435 | void SDMath::multiplyInSitu(SDMemTable* in, Float factor) {
|
---|
| 436 | SDMemTable* sdmt = new SDMemTable(*in);
|
---|
| 437 | Table t = sdmt->table();
|
---|
| 438 | ArrayColumn<Float> spec(t,"SPECTRA");
|
---|
| 439 | for (uInt i=0; i < t.nrow(); i++) {
|
---|
| 440 | MaskedArray<Float> marr(sdmt->rowAsMaskedArray(i));
|
---|
| 441 | marr *= factor;
|
---|
| 442 | spec.put(i, marr.getArray());
|
---|
| 443 | }
|
---|
| 444 | in = sdmt;
|
---|
| 445 | delete sdmt;sdmt=0;
|
---|
| 446 | }
|
---|
| 447 |
|
---|
[85] | 448 | CountedPtr<SDMemTable>
|
---|
[130] | 449 | SDMath::multiply(const CountedPtr<SDMemTable>& in, Float factor)
|
---|
| 450 | //
|
---|
| 451 | // Multiply values by factor
|
---|
| 452 | //
|
---|
| 453 | {
|
---|
[15] | 454 | SDMemTable* sdmt = new SDMemTable(*in);
|
---|
| 455 | Table t = sdmt->table();
|
---|
| 456 | ArrayColumn<Float> spec(t,"SPECTRA");
|
---|
[9] | 457 |
|
---|
[15] | 458 | for (uInt i=0; i < t.nrow(); i++) {
|
---|
| 459 | MaskedArray<Float> marr(sdmt->rowAsMaskedArray(i));
|
---|
| 460 | marr *= factor;
|
---|
| 461 | spec.put(i, marr.getArray());
|
---|
| 462 | }
|
---|
| 463 | return CountedPtr<SDMemTable>(sdmt);
|
---|
| 464 | }
|
---|
[48] | 465 |
|
---|
[107] | 466 | CountedPtr<SDMemTable>
|
---|
[130] | 467 | SDMath::add(const CountedPtr<SDMemTable>& in, Float offset)
|
---|
| 468 | //
|
---|
| 469 | // Add offset to values
|
---|
| 470 | //
|
---|
| 471 | {
|
---|
[107] | 472 | SDMemTable* sdmt = new SDMemTable(*in);
|
---|
[130] | 473 |
|
---|
[107] | 474 | Table t = sdmt->table();
|
---|
| 475 | ArrayColumn<Float> spec(t,"SPECTRA");
|
---|
| 476 |
|
---|
| 477 | for (uInt i=0; i < t.nrow(); i++) {
|
---|
| 478 | MaskedArray<Float> marr(sdmt->rowAsMaskedArray(i));
|
---|
| 479 | marr += offset;
|
---|
| 480 | spec.put(i, marr.getArray());
|
---|
| 481 | }
|
---|
| 482 | return CountedPtr<SDMemTable>(sdmt);
|
---|
| 483 | }
|
---|
| 484 |
|
---|
| 485 |
|
---|
[130] | 486 | CountedPtr<SDMemTable>
|
---|
| 487 | SDMath::hanning(const CountedPtr<SDMemTable>& in)
|
---|
| 488 | //
|
---|
| 489 | // Hanning smooth each row
|
---|
| 490 | // Should Tsys be smoothed ?
|
---|
| 491 | //
|
---|
| 492 | {
|
---|
| 493 | SDMemTable* sdmt = new SDMemTable(*in,True);
|
---|
[107] | 494 |
|
---|
[130] | 495 | // Loop over rows in Table
|
---|
[48] | 496 |
|
---|
[130] | 497 | for (uInt ri=0; ri < in->nRow(); ++ri) {
|
---|
[38] | 498 |
|
---|
[130] | 499 | // Get data
|
---|
[125] | 500 |
|
---|
[130] | 501 | const MaskedArray<Float>& marr(in->rowAsMaskedArray(ri));
|
---|
[85] | 502 | Array<Float> arr = marr.getArray();
|
---|
| 503 | Array<Bool> barr = marr.getMask();
|
---|
[130] | 504 |
|
---|
| 505 | // Smooth along the channels axis
|
---|
| 506 |
|
---|
| 507 | uInt axis = 3;
|
---|
| 508 | VectorIterator<Float> itData(arr, axis);
|
---|
| 509 | VectorIterator<Bool> itMask(barr, axis);
|
---|
| 510 | Vector<Float> outv;
|
---|
| 511 | Vector<Bool> outm;
|
---|
| 512 | while (!itData.pastEnd()) {
|
---|
| 513 | mathutil::hanning(outv, outm, itData.vector(), itMask.vector());
|
---|
| 514 | itData.vector() = outv;
|
---|
| 515 | itMask.vector() = outm;
|
---|
| 516 | //
|
---|
| 517 | itData.next();
|
---|
| 518 | itMask.next();
|
---|
[38] | 519 | }
|
---|
[130] | 520 |
|
---|
| 521 | // Create and put back
|
---|
| 522 |
|
---|
[85] | 523 | Array<uChar> outflags(barr.shape());
|
---|
| 524 | convertArray(outflags,!barr);
|
---|
| 525 | SDContainer sc = in->getSDContainer(ri);
|
---|
| 526 | sc.putSpectrum(arr);
|
---|
| 527 | sc.putFlags(outflags);
|
---|
| 528 | sdmt->putSDContainer(sc);
|
---|
[38] | 529 | }
|
---|
| 530 | return CountedPtr<SDMemTable>(sdmt);
|
---|
| 531 | }
|
---|
| 532 |
|
---|
[85] | 533 |
|
---|
[130] | 534 |
|
---|
| 535 |
|
---|
[85] | 536 | CountedPtr<SDMemTable>
|
---|
| 537 | SDMath::averagePol(const CountedPtr<SDMemTable>& in,
|
---|
[130] | 538 | const Vector<Bool>& mask)
|
---|
| 539 | {
|
---|
| 540 | const uInt nRows = in->nRow();
|
---|
| 541 | const uInt axis = 3; // Spectrum
|
---|
| 542 | const IPosition axes(2, 2, 3); // pol-channel plane
|
---|
| 543 |
|
---|
| 544 | // Create output Table
|
---|
| 545 |
|
---|
| 546 | SDMemTable* sdmt = new SDMemTable(*in, True);
|
---|
| 547 |
|
---|
| 548 | // Loop over rows
|
---|
| 549 |
|
---|
| 550 | for (uInt iRow=0; iRow<nRows; iRow++) {
|
---|
| 551 |
|
---|
| 552 | // Get data for this row
|
---|
| 553 |
|
---|
| 554 | MaskedArray<Float> marr(in->rowAsMaskedArray(iRow));
|
---|
| 555 | Array<Float>& arr = marr.getRWArray();
|
---|
| 556 | const Array<Bool>& barr = marr.getMask();
|
---|
| 557 | //
|
---|
| 558 | IPosition shp = marr.shape();
|
---|
| 559 | const Bool useMask = (mask.nelements() == shp(axis));
|
---|
| 560 | const uInt nChan = shp(axis);
|
---|
| 561 |
|
---|
| 562 | // Make iterators to iterate by pol-channel planes
|
---|
| 563 |
|
---|
| 564 | ArrayIterator<Float> itDataPlane(arr, axes);
|
---|
| 565 | ReadOnlyArrayIterator<Bool> itMaskPlane(barr, axes);
|
---|
| 566 |
|
---|
| 567 | // Accumulations
|
---|
| 568 |
|
---|
| 569 | Float fac = 0.0;
|
---|
| 570 | Vector<Float> vecSum(nChan,0.0);
|
---|
| 571 |
|
---|
| 572 | // Iterate by plane
|
---|
| 573 |
|
---|
| 574 | while (!itDataPlane.pastEnd()) {
|
---|
| 575 |
|
---|
| 576 | // Iterate through pol-channel plane by spectrum
|
---|
| 577 |
|
---|
| 578 | Vector<Float> t1(nChan); t1 = 0.0;
|
---|
| 579 | Vector<Bool> t2(nChan); t2 = True;
|
---|
| 580 | MaskedArray<Float> vecSum(t1,t2);
|
---|
| 581 | Float varSum = 0.0;
|
---|
| 582 | {
|
---|
| 583 | ReadOnlyVectorIterator<Float> itDataVec(itDataPlane.array(), 1);
|
---|
| 584 | ReadOnlyVectorIterator<Bool> itMaskVec(itMaskPlane.array(), 1);
|
---|
| 585 | while (!itDataVec.pastEnd()) {
|
---|
| 586 |
|
---|
| 587 | // Create MA of data & mask (optionally including OTF mask) and get variance
|
---|
| 588 |
|
---|
| 589 | if (useMask) {
|
---|
| 590 | const MaskedArray<Float> spec(itDataVec.vector(),mask&&itMaskVec.vector());
|
---|
| 591 | fac = 1.0 / variance(spec);
|
---|
| 592 | } else {
|
---|
| 593 | const MaskedArray<Float> spec(itDataVec.vector(),itMaskVec.vector());
|
---|
| 594 | fac = 1.0 / variance(spec);
|
---|
| 595 | }
|
---|
| 596 |
|
---|
| 597 | // Normalize spectrum (without OTF mask) and accumulate
|
---|
| 598 |
|
---|
| 599 | const MaskedArray<Float> spec(fac*itDataVec.vector(), itMaskVec.vector());
|
---|
| 600 | vecSum += spec;
|
---|
| 601 | varSum += fac;
|
---|
| 602 |
|
---|
| 603 | // Next
|
---|
| 604 |
|
---|
| 605 | itDataVec.next();
|
---|
| 606 | itMaskVec.next();
|
---|
| 607 | }
|
---|
[85] | 608 | }
|
---|
[48] | 609 |
|
---|
[130] | 610 | // Normalize summed spectrum
|
---|
[48] | 611 |
|
---|
[130] | 612 | vecSum /= varSum;
|
---|
[48] | 613 |
|
---|
[130] | 614 | // We have formed the weighted averaged spectrum from all polarizations
|
---|
| 615 | // for this beam and IF. Now replicate the spectrum to all polarizations
|
---|
[48] | 616 |
|
---|
[130] | 617 | {
|
---|
| 618 | VectorIterator<Float> itDataVec(itDataPlane.array(), 1); // Writes back into 'arr'
|
---|
| 619 | const Vector<Float>& vecSumData = vecSum.getArray(); // It *is* a Vector
|
---|
| 620 | //
|
---|
| 621 | while (!itDataVec.pastEnd()) {
|
---|
| 622 | itDataVec.vector() = vecSumData;
|
---|
| 623 | itDataVec.next();
|
---|
| 624 | }
|
---|
| 625 | }
|
---|
| 626 |
|
---|
| 627 | // Step to next beam/IF combination
|
---|
| 628 |
|
---|
| 629 | itDataPlane.next();
|
---|
| 630 | itMaskPlane.next();
|
---|
| 631 | }
|
---|
| 632 |
|
---|
| 633 | // Generate output container and write it to output table
|
---|
| 634 |
|
---|
| 635 | SDContainer sc = in->getSDContainer();
|
---|
| 636 | Array<uChar> outflags(barr.shape());
|
---|
| 637 | convertArray(outflags,!barr);
|
---|
| 638 | sc.putSpectrum(arr);
|
---|
| 639 | sc.putFlags(outflags);
|
---|
| 640 | sdmt->putSDContainer(sc);
|
---|
| 641 | }
|
---|
| 642 | //
|
---|
| 643 | return CountedPtr<SDMemTable>(sdmt);
|
---|
[48] | 644 | }
|
---|
| 645 |
|
---|
[130] | 646 |
|
---|
[85] | 647 | CountedPtr<SDMemTable> SDMath::bin(const CountedPtr<SDMemTable>& in,
|
---|
[130] | 648 | Int width)
|
---|
| 649 | {
|
---|
[48] | 650 | SDHeader sh = in->getSDHeader();
|
---|
[85] | 651 | SDMemTable* sdmt = new SDMemTable(*in,True);
|
---|
| 652 |
|
---|
[130] | 653 | // Bin up SpectralCoordinates
|
---|
| 654 |
|
---|
| 655 | IPosition factors(1);
|
---|
| 656 | factors(0) = width;
|
---|
[85] | 657 | for (uInt j=0; j<in->nCoordinates(); ++j) {
|
---|
[130] | 658 | CoordinateSystem cSys;
|
---|
| 659 | cSys.addCoordinate(in->getCoordinate(j));
|
---|
| 660 | CoordinateSystem cSysBin =
|
---|
| 661 | CoordinateUtil::makeBinnedCoordinateSystem (factors, cSys, False);
|
---|
| 662 | //
|
---|
| 663 | SpectralCoordinate sCBin = cSysBin.spectralCoordinate(0);
|
---|
| 664 | sdmt->setCoordinate(sCBin, j);
|
---|
[85] | 665 | }
|
---|
[130] | 666 |
|
---|
| 667 | // Use RebinLattice to find shape
|
---|
| 668 |
|
---|
| 669 | IPosition shapeIn(1,sh.nchan);
|
---|
| 670 | IPosition shapeOut = RebinLattice<Float>::rebinShape (shapeIn, factors);
|
---|
| 671 | sh.nchan = shapeOut(0);
|
---|
[48] | 672 | sdmt->putSDHeader(sh);
|
---|
[85] | 673 |
|
---|
[130] | 674 |
|
---|
| 675 | // Loop over rows and bin along channel axis
|
---|
| 676 |
|
---|
| 677 | const uInt axis = 3;
|
---|
[85] | 678 | for (uInt i=0; i < in->nRow(); ++i) {
|
---|
[130] | 679 | SDContainer sc = in->getSDContainer(i);
|
---|
| 680 | //
|
---|
| 681 | Array<Float> tSys(sc.getTsys()); // Get it out before sc changes shape
|
---|
| 682 |
|
---|
| 683 | // Bin up spectrum
|
---|
| 684 |
|
---|
[85] | 685 | MaskedArray<Float> marr(in->rowAsMaskedArray(i));
|
---|
| 686 | MaskedArray<Float> marrout;
|
---|
[130] | 687 | LatticeUtilities::bin(marrout, marr, axis, width);
|
---|
| 688 |
|
---|
| 689 | // Put back the binned data and flags
|
---|
| 690 |
|
---|
[85] | 691 | IPosition ip2 = marrout.shape();
|
---|
| 692 | sc.resize(ip2);
|
---|
| 693 | sc.putSpectrum(marrout.getArray());
|
---|
[130] | 694 | //
|
---|
[85] | 695 | Array<uChar> outflags(ip2);
|
---|
| 696 | convertArray(outflags,!(marrout.getMask()));
|
---|
| 697 | sc.putFlags(outflags);
|
---|
[130] | 698 |
|
---|
| 699 | // Bin up Tsys.
|
---|
| 700 |
|
---|
| 701 | Array<Bool> allGood(tSys.shape(),True);
|
---|
| 702 | MaskedArray<Float> tSysIn(tSys, allGood, True);
|
---|
| 703 | //
|
---|
| 704 | MaskedArray<Float> tSysOut;
|
---|
| 705 | LatticeUtilities::bin(tSysOut, tSysIn, axis, width);
|
---|
| 706 | sc.putTsys(tSysOut.getArray());
|
---|
[85] | 707 | sdmt->putSDContainer(sc);
|
---|
| 708 | }
|
---|
[48] | 709 | return CountedPtr<SDMemTable>(sdmt);
|
---|
| 710 | }
|
---|
[130] | 711 |
|
---|
| 712 |
|
---|
| 713 |
|
---|
| 714 | std::vector<float> SDMath::statistic (const CountedPtr<SDMemTable>& in,
|
---|
| 715 | const std::vector<bool>& mask,
|
---|
| 716 | const std::string& which)
|
---|
| 717 | //
|
---|
| 718 | // Perhaps iteration over pol/beam/if should be in here
|
---|
| 719 | // and inside the nrow iteration ?
|
---|
| 720 | //
|
---|
| 721 | {
|
---|
| 722 | const uInt nRow = in->nRow();
|
---|
| 723 | std::vector<float> result(nRow);
|
---|
| 724 | Vector<Bool> msk(mask);
|
---|
| 725 |
|
---|
| 726 | // Specify cursor location
|
---|
| 727 |
|
---|
| 728 | uInt i = in->getBeam();
|
---|
| 729 | uInt j = in->getIF();
|
---|
| 730 | uInt k = in->getPol();
|
---|
| 731 | IPosition start(4,i,j,k,0);
|
---|
| 732 | IPosition end(4,i,j,k,in->nChan()-1);
|
---|
| 733 |
|
---|
| 734 | // Loop over rows
|
---|
| 735 |
|
---|
| 736 | const uInt nEl = msk.nelements();
|
---|
| 737 | for (uInt ii=0; ii < in->nRow(); ++ii) {
|
---|
| 738 |
|
---|
| 739 | // Get row and deconstruct
|
---|
| 740 |
|
---|
| 741 | MaskedArray<Float> marr(in->rowAsMaskedArray(ii));
|
---|
| 742 | Array<Float> arr = marr.getArray();
|
---|
| 743 | Array<Bool> barr = marr.getMask();
|
---|
| 744 |
|
---|
| 745 | // Access desired piece of data
|
---|
| 746 |
|
---|
| 747 | Array<Float> v((arr(start,end)).nonDegenerate());
|
---|
| 748 | Array<Bool> m((barr(start,end)).nonDegenerate());
|
---|
| 749 |
|
---|
| 750 | // Apply OTF mask
|
---|
| 751 |
|
---|
| 752 | MaskedArray<Float> tmp;
|
---|
| 753 | if (m.nelements()==nEl) {
|
---|
| 754 | tmp.setData(v,m&&msk);
|
---|
| 755 | } else {
|
---|
| 756 | tmp.setData(v,m);
|
---|
| 757 | }
|
---|
| 758 |
|
---|
| 759 | // Get statistic
|
---|
| 760 |
|
---|
[144] | 761 | result[ii] = mathutil::statistics(which, tmp);
|
---|
[130] | 762 | }
|
---|
| 763 | //
|
---|
| 764 | return result;
|
---|
| 765 | }
|
---|
| 766 |
|
---|
[144] | 767 | void SDMath::fillSDC (SDContainer& sc,
|
---|
| 768 | const Array<Bool>& mask,
|
---|
| 769 | const Array<Float>& data,
|
---|
| 770 | const Array<Float>& tSys,
|
---|
| 771 | Int scanID, Double timeStamp,
|
---|
| 772 | Double interval, const String& sourceName,
|
---|
| 773 | const Vector<uInt>& freqID)
|
---|
| 774 | {
|
---|
| 775 | sc.putSpectrum(data);
|
---|
| 776 | //
|
---|
| 777 | Array<uChar> outflags(mask.shape());
|
---|
| 778 | convertArray(outflags,!mask);
|
---|
| 779 | sc.putFlags(outflags);
|
---|
| 780 | //
|
---|
| 781 | sc.putTsys(tSys);
|
---|
[130] | 782 |
|
---|
[144] | 783 | // Time things
|
---|
| 784 |
|
---|
| 785 | sc.timestamp = timeStamp;
|
---|
| 786 | sc.interval = interval;
|
---|
| 787 | sc.scanid = scanID;
|
---|
| 788 | //
|
---|
| 789 | sc.sourcename = sourceName;
|
---|
| 790 | sc.putFreqMap(freqID);
|
---|
| 791 | }
|
---|
| 792 |
|
---|
| 793 | void SDMath::normalize (MaskedArray<Float>& sum,
|
---|
| 794 | const Array<Float>& sumSq,
|
---|
| 795 | const Array<Float>& nPts,
|
---|
| 796 | weightType wtType, Int axis,
|
---|
| 797 | Int nAxesSub)
|
---|
[130] | 798 | {
|
---|
[144] | 799 | IPosition pos2(nAxesSub,0);
|
---|
| 800 | //
|
---|
| 801 | if (wtType==NONE) {
|
---|
| 802 |
|
---|
| 803 | // We just average by the number of points accumulated.
|
---|
| 804 | // We need to make a MA out of nPts so that no divide by
|
---|
| 805 | // zeros occur
|
---|
| 806 |
|
---|
| 807 | MaskedArray<Float> t(nPts, (nPts>Float(0.0)));
|
---|
| 808 | sum /= t;
|
---|
| 809 | } else if (wtType==VAR) {
|
---|
| 810 |
|
---|
| 811 | // Normalize each spectrum by sum(1/var) where the variance
|
---|
| 812 | // is worked out for each spectrum
|
---|
| 813 |
|
---|
| 814 | Array<Float>& data = sum.getRWArray();
|
---|
| 815 | VectorIterator<Float> itData(data, axis);
|
---|
| 816 | while (!itData.pastEnd()) {
|
---|
| 817 | pos2 = itData.pos().getFirst(nAxesSub);
|
---|
| 818 | itData.vector() /= sumSq(pos2);
|
---|
| 819 | itData.next();
|
---|
| 820 | }
|
---|
| 821 | } else if (wtType==TSYS) {
|
---|
[130] | 822 | }
|
---|
| 823 | }
|
---|
[144] | 824 |
|
---|