[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> |
---|
[177] | 44 | #include <casa/Utilities/Assert.h> |
---|
[130] | 45 | #include <casa/Exceptions.h> |
---|
[2] | 46 | |
---|
[177] | 47 | #include <scimath/Mathematics/VectorKernel.h> |
---|
| 48 | #include <scimath/Mathematics/Convolver.h> |
---|
| 49 | |
---|
[81] | 50 | #include <tables/Tables/Table.h> |
---|
| 51 | #include <tables/Tables/ScalarColumn.h> |
---|
| 52 | #include <tables/Tables/ArrayColumn.h> |
---|
[2] | 53 | |
---|
[130] | 54 | #include <lattices/Lattices/LatticeUtilities.h> |
---|
| 55 | #include <lattices/Lattices/RebinLattice.h> |
---|
[81] | 56 | #include <coordinates/Coordinates/SpectralCoordinate.h> |
---|
[130] | 57 | #include <coordinates/Coordinates/CoordinateSystem.h> |
---|
| 58 | #include <coordinates/Coordinates/CoordinateUtil.h> |
---|
[38] | 59 | |
---|
| 60 | #include "MathUtils.h" |
---|
[209] | 61 | #include "Definitions.h" |
---|
[2] | 62 | #include "SDContainer.h" |
---|
| 63 | #include "SDMemTable.h" |
---|
| 64 | |
---|
| 65 | #include "SDMath.h" |
---|
| 66 | |
---|
[125] | 67 | using namespace casa; |
---|
[83] | 68 | using namespace asap; |
---|
[2] | 69 | |
---|
[170] | 70 | |
---|
| 71 | SDMath::SDMath() |
---|
| 72 | {;} |
---|
| 73 | |
---|
[185] | 74 | SDMath::SDMath(const SDMath& other) |
---|
[170] | 75 | { |
---|
| 76 | |
---|
| 77 | // No state |
---|
| 78 | |
---|
| 79 | } |
---|
| 80 | |
---|
| 81 | SDMath& SDMath::operator=(const SDMath& other) |
---|
| 82 | { |
---|
| 83 | if (this != &other) { |
---|
| 84 | // No state |
---|
| 85 | } |
---|
| 86 | return *this; |
---|
| 87 | } |
---|
| 88 | |
---|
[183] | 89 | SDMath::~SDMath() |
---|
| 90 | {;} |
---|
[170] | 91 | |
---|
[183] | 92 | |
---|
[185] | 93 | CountedPtr<SDMemTable> SDMath::average(const Block<CountedPtr<SDMemTable> >& in, |
---|
| 94 | const Vector<Bool>& mask, Bool scanAv, |
---|
| 95 | const std::string& weightStr) |
---|
[130] | 96 | // |
---|
[144] | 97 | // Weighted averaging of spectra from one or more Tables. |
---|
[130] | 98 | // |
---|
| 99 | { |
---|
[2] | 100 | |
---|
[163] | 101 | // Convert weight type |
---|
| 102 | |
---|
| 103 | WeightType wtType = NONE; |
---|
[185] | 104 | convertWeightString(wtType, weightStr); |
---|
[163] | 105 | |
---|
[144] | 106 | // Create output Table by cloning from the first table |
---|
[2] | 107 | |
---|
[144] | 108 | SDMemTable* pTabOut = new SDMemTable(*in[0],True); |
---|
[130] | 109 | |
---|
[144] | 110 | // Setup |
---|
[130] | 111 | |
---|
[209] | 112 | const uInt axis = asap::ChanAxis; // Spectral axis |
---|
[144] | 113 | IPosition shp = in[0]->rowAsMaskedArray(0).shape(); // Must not change |
---|
| 114 | Array<Float> arr(shp); |
---|
| 115 | Array<Bool> barr(shp); |
---|
| 116 | const Bool useMask = (mask.nelements() == shp(axis)); |
---|
[130] | 117 | |
---|
[144] | 118 | // Columns from Tables |
---|
[130] | 119 | |
---|
[144] | 120 | ROArrayColumn<Float> tSysCol; |
---|
| 121 | ROScalarColumn<Double> mjdCol; |
---|
| 122 | ROScalarColumn<String> srcNameCol; |
---|
| 123 | ROScalarColumn<Double> intCol; |
---|
| 124 | ROArrayColumn<uInt> fqIDCol; |
---|
[130] | 125 | |
---|
[144] | 126 | // Create accumulation MaskedArray. We accumulate for each channel,if,pol,beam |
---|
| 127 | // Note that the mask of the accumulation array will ALWAYS remain ALL True. |
---|
| 128 | // The MA is only used so that when data which is masked Bad is added to it, |
---|
| 129 | // that data does not contribute. |
---|
| 130 | |
---|
| 131 | Array<Float> zero(shp); |
---|
| 132 | zero=0.0; |
---|
| 133 | Array<Bool> good(shp); |
---|
| 134 | good = True; |
---|
| 135 | MaskedArray<Float> sum(zero,good); |
---|
| 136 | |
---|
| 137 | // Counter arrays |
---|
| 138 | |
---|
| 139 | Array<Float> nPts(shp); // Number of points |
---|
| 140 | nPts = 0.0; |
---|
| 141 | Array<Float> nInc(shp); // Increment |
---|
| 142 | nInc = 1.0; |
---|
| 143 | |
---|
| 144 | // Create accumulation Array for variance. We accumulate for |
---|
| 145 | // each if,pol,beam, but average over channel. So we need |
---|
| 146 | // a shape with one less axis dropping channels. |
---|
| 147 | |
---|
| 148 | const uInt nAxesSub = shp.nelements() - 1; |
---|
| 149 | IPosition shp2(nAxesSub); |
---|
| 150 | for (uInt i=0,j=0; i<(nAxesSub+1); i++) { |
---|
| 151 | if (i!=axis) { |
---|
| 152 | shp2(j) = shp(i); |
---|
| 153 | j++; |
---|
| 154 | } |
---|
[2] | 155 | } |
---|
[144] | 156 | Array<Float> sumSq(shp2); |
---|
| 157 | sumSq = 0.0; |
---|
| 158 | IPosition pos2(nAxesSub,0); // For indexing |
---|
[130] | 159 | |
---|
[144] | 160 | // Time-related accumulators |
---|
[130] | 161 | |
---|
[144] | 162 | Double time; |
---|
| 163 | Double timeSum = 0.0; |
---|
| 164 | Double intSum = 0.0; |
---|
| 165 | Double interval = 0.0; |
---|
[130] | 166 | |
---|
[144] | 167 | // To get the right shape for the Tsys accumulator we need to |
---|
| 168 | // access a column from the first table. The shape of this |
---|
| 169 | // array must not change |
---|
[130] | 170 | |
---|
[144] | 171 | Array<Float> tSysSum; |
---|
| 172 | { |
---|
| 173 | const Table& tabIn = in[0]->table(); |
---|
| 174 | tSysCol.attach(tabIn,"TSYS"); |
---|
| 175 | tSysSum.resize(tSysCol.shape(0)); |
---|
| 176 | } |
---|
| 177 | tSysSum =0.0; |
---|
| 178 | Array<Float> tSys; |
---|
| 179 | |
---|
| 180 | // Scan and row tracking |
---|
| 181 | |
---|
| 182 | Int oldScanID = 0; |
---|
| 183 | Int outScanID = 0; |
---|
| 184 | Int scanID = 0; |
---|
| 185 | Int rowStart = 0; |
---|
| 186 | Int nAccum = 0; |
---|
| 187 | Int tableStart = 0; |
---|
| 188 | |
---|
| 189 | // Source and FreqID |
---|
| 190 | |
---|
| 191 | String sourceName, oldSourceName, sourceNameStart; |
---|
| 192 | Vector<uInt> freqID, freqIDStart, oldFreqID; |
---|
| 193 | |
---|
| 194 | // Loop over tables |
---|
| 195 | |
---|
| 196 | Float fac = 1.0; |
---|
| 197 | const uInt nTables = in.nelements(); |
---|
| 198 | for (uInt iTab=0; iTab<nTables; iTab++) { |
---|
| 199 | |
---|
| 200 | // Attach columns to Table |
---|
| 201 | |
---|
| 202 | const Table& tabIn = in[iTab]->table(); |
---|
| 203 | tSysCol.attach(tabIn, "TSYS"); |
---|
| 204 | mjdCol.attach(tabIn, "TIME"); |
---|
| 205 | srcNameCol.attach(tabIn, "SRCNAME"); |
---|
| 206 | intCol.attach(tabIn, "INTERVAL"); |
---|
| 207 | fqIDCol.attach(tabIn, "FREQID"); |
---|
| 208 | |
---|
| 209 | // Loop over rows in Table |
---|
| 210 | |
---|
| 211 | const uInt nRows = in[iTab]->nRow(); |
---|
| 212 | for (uInt iRow=0; iRow<nRows; iRow++) { |
---|
| 213 | |
---|
| 214 | // Check conformance |
---|
| 215 | |
---|
| 216 | IPosition shp2 = in[iTab]->rowAsMaskedArray(iRow).shape(); |
---|
| 217 | if (!shp.isEqual(shp2)) { |
---|
| 218 | throw (AipsError("Shapes for all rows must be the same")); |
---|
| 219 | } |
---|
| 220 | |
---|
| 221 | // If we are not doing scan averages, make checks for source and |
---|
| 222 | // frequency setup and warn if averaging across them |
---|
| 223 | |
---|
| 224 | // Get copy of Scan Container for this row |
---|
| 225 | |
---|
| 226 | SDContainer sc = in[iTab]->getSDContainer(iRow); |
---|
| 227 | scanID = sc.scanid; |
---|
| 228 | |
---|
| 229 | // Get quantities from columns |
---|
| 230 | |
---|
| 231 | srcNameCol.getScalar(iRow, sourceName); |
---|
| 232 | mjdCol.get(iRow, time); |
---|
| 233 | tSysCol.get(iRow, tSys); |
---|
| 234 | intCol.get(iRow, interval); |
---|
| 235 | fqIDCol.get(iRow, freqID); |
---|
| 236 | |
---|
| 237 | // Initialize first source and freqID |
---|
| 238 | |
---|
| 239 | if (iRow==0 && iTab==0) { |
---|
| 240 | sourceNameStart = sourceName; |
---|
| 241 | freqIDStart = freqID; |
---|
| 242 | } |
---|
| 243 | |
---|
| 244 | // If we are doing scan averages, see if we are at the end of an |
---|
| 245 | // accumulation period (scan). We must check soutce names too, |
---|
| 246 | // since we might have two tables with one scan each but different |
---|
| 247 | // source names; we shouldn't average different sources together |
---|
| 248 | |
---|
| 249 | if (scanAv && ( (scanID != oldScanID) || |
---|
| 250 | (iRow==0 && iTab>0 && sourceName!=oldSourceName))) { |
---|
| 251 | |
---|
| 252 | // Normalize data in 'sum' accumulation array according to weighting scheme |
---|
| 253 | |
---|
[185] | 254 | normalize(sum, sumSq, nPts, wtType, axis, nAxesSub); |
---|
[144] | 255 | |
---|
| 256 | // Fill scan container. The source and freqID come from the |
---|
| 257 | // first row of the first table that went into this average ( |
---|
| 258 | // should be the same for all rows in the scan average) |
---|
| 259 | |
---|
| 260 | Float nR(nAccum); |
---|
[185] | 261 | fillSDC(sc, sum.getMask(), sum.getArray(), tSysSum/nR, outScanID, |
---|
[144] | 262 | timeSum/nR, intSum, sourceNameStart, freqIDStart); |
---|
| 263 | |
---|
| 264 | // Write container out to Table |
---|
| 265 | |
---|
| 266 | pTabOut->putSDContainer(sc); |
---|
| 267 | |
---|
| 268 | // Reset accumulators |
---|
| 269 | |
---|
| 270 | sum = 0.0; |
---|
| 271 | sumSq = 0.0; |
---|
| 272 | nAccum = 0; |
---|
| 273 | // |
---|
| 274 | tSysSum =0.0; |
---|
| 275 | timeSum = 0.0; |
---|
| 276 | intSum = 0.0; |
---|
[184] | 277 | nPts = 0.0; // reset this too!!! |
---|
[144] | 278 | |
---|
| 279 | // Increment |
---|
| 280 | |
---|
| 281 | rowStart = iRow; // First row for next accumulation |
---|
| 282 | tableStart = iTab; // First table for next accumulation |
---|
| 283 | sourceNameStart = sourceName; // First source name for next accumulation |
---|
| 284 | freqIDStart = freqID; // First FreqID for next accumulation |
---|
| 285 | // |
---|
| 286 | oldScanID = scanID; |
---|
| 287 | outScanID += 1; // Scan ID for next accumulation period |
---|
| 288 | |
---|
[184] | 289 | } |
---|
| 290 | |
---|
[146] | 291 | // Accumulate |
---|
[144] | 292 | |
---|
[185] | 293 | accumulate(timeSum, intSum, nAccum, sum, sumSq, nPts, tSysSum, |
---|
[146] | 294 | tSys, nInc, mask, time, interval, in, iTab, iRow, axis, |
---|
| 295 | nAxesSub, useMask, wtType); |
---|
[144] | 296 | // |
---|
| 297 | oldSourceName = sourceName; |
---|
| 298 | oldFreqID = freqID; |
---|
[184] | 299 | } |
---|
[144] | 300 | } |
---|
| 301 | |
---|
| 302 | // OK at this point we have accumulation data which is either |
---|
| 303 | // - accumulated from all tables into one row |
---|
| 304 | // or |
---|
| 305 | // - accumulated from the last scan average |
---|
| 306 | // |
---|
| 307 | // Normalize data in 'sum' accumulation array according to weighting scheme |
---|
[185] | 308 | normalize(sum, sumSq, nPts, wtType, axis, nAxesSub); |
---|
[144] | 309 | |
---|
| 310 | // Create and fill container. The container we clone will be from |
---|
| 311 | // the last Table and the first row that went into the current |
---|
| 312 | // accumulation. It probably doesn't matter that much really... |
---|
| 313 | |
---|
| 314 | Float nR(nAccum); |
---|
| 315 | SDContainer sc = in[tableStart]->getSDContainer(rowStart); |
---|
[185] | 316 | fillSDC(sc, sum.getMask(), sum.getArray(), tSysSum/nR, outScanID, |
---|
[144] | 317 | timeSum/nR, intSum, sourceNameStart, freqIDStart); |
---|
| 318 | // |
---|
| 319 | pTabOut->putSDContainer(sc); |
---|
| 320 | /* |
---|
| 321 | cout << endl; |
---|
| 322 | cout << "Last accumulation for output scan ID " << outScanID << endl; |
---|
| 323 | cout << " The first row in this accumulation is " << rowStart << endl; |
---|
| 324 | cout << " The number of rows accumulated is " << nAccum << endl; |
---|
| 325 | cout << " The first table in this accumulation is " << tableStart << endl; |
---|
| 326 | cout << " The first source in this accumulation is " << sourceNameStart << endl; |
---|
| 327 | cout << " The first freqID in this accumulation is " << freqIDStart << endl; |
---|
| 328 | cout << " Average time stamp = " << timeSum/nR << endl; |
---|
| 329 | cout << " Integrated time = " << intSum << endl; |
---|
| 330 | */ |
---|
| 331 | return CountedPtr<SDMemTable>(pTabOut); |
---|
[2] | 332 | } |
---|
[9] | 333 | |
---|
[144] | 334 | |
---|
| 335 | |
---|
[85] | 336 | CountedPtr<SDMemTable> |
---|
| 337 | SDMath::quotient(const CountedPtr<SDMemTable>& on, |
---|
[185] | 338 | const CountedPtr<SDMemTable>& off) |
---|
| 339 | { |
---|
[130] | 340 | // |
---|
| 341 | // Compute quotient spectrum |
---|
| 342 | // |
---|
| 343 | const uInt nRows = on->nRow(); |
---|
| 344 | if (off->nRow() != nRows) { |
---|
| 345 | throw (AipsError("Input Scan Tables must have the same number of rows")); |
---|
| 346 | } |
---|
[85] | 347 | |
---|
[130] | 348 | // Input Tables and columns |
---|
| 349 | |
---|
[9] | 350 | Table ton = on->table(); |
---|
| 351 | Table toff = off->table(); |
---|
[85] | 352 | ROArrayColumn<Float> tsys(toff, "TSYS"); |
---|
[9] | 353 | ROScalarColumn<Double> mjd(ton, "TIME"); |
---|
[15] | 354 | ROScalarColumn<Double> integr(ton, "INTERVAL"); |
---|
[9] | 355 | ROScalarColumn<String> srcn(ton, "SRCNAME"); |
---|
[38] | 356 | ROArrayColumn<uInt> freqidc(ton, "FREQID"); |
---|
| 357 | |
---|
[130] | 358 | // Output Table cloned from input |
---|
[85] | 359 | |
---|
[171] | 360 | SDMemTable* pTabOut = new SDMemTable(*on, True); |
---|
[130] | 361 | |
---|
| 362 | // Loop over rows |
---|
| 363 | |
---|
| 364 | for (uInt i=0; i<nRows; i++) { |
---|
| 365 | MaskedArray<Float> mon(on->rowAsMaskedArray(i)); |
---|
| 366 | MaskedArray<Float> moff(off->rowAsMaskedArray(i)); |
---|
| 367 | IPosition ipon = mon.shape(); |
---|
| 368 | IPosition ipoff = moff.shape(); |
---|
| 369 | // |
---|
| 370 | Array<Float> tsarr; |
---|
| 371 | tsys.get(i, tsarr); |
---|
| 372 | if (ipon != ipoff && ipon != tsarr.shape()) { |
---|
| 373 | throw(AipsError("on/off not conformant")); |
---|
| 374 | } |
---|
| 375 | |
---|
| 376 | // Compute quotient |
---|
| 377 | |
---|
| 378 | MaskedArray<Float> tmp = (mon-moff); |
---|
| 379 | Array<Float> out(tmp.getArray()); |
---|
| 380 | out /= moff; |
---|
| 381 | out *= tsarr; |
---|
[163] | 382 | Array<Bool> outflagsb = mon.getMask() && moff.getMask(); |
---|
[130] | 383 | |
---|
| 384 | // Fill container for this row |
---|
| 385 | |
---|
[184] | 386 | SDContainer sc = on->getSDContainer(i); |
---|
[163] | 387 | // |
---|
[185] | 388 | putDataInSDC(sc, out, outflagsb); |
---|
[130] | 389 | sc.putTsys(tsarr); |
---|
[184] | 390 | sc.scanid = i; |
---|
[130] | 391 | |
---|
| 392 | // Put new row in output Table |
---|
| 393 | |
---|
[171] | 394 | pTabOut->putSDContainer(sc); |
---|
[130] | 395 | } |
---|
| 396 | // |
---|
[171] | 397 | return CountedPtr<SDMemTable>(pTabOut); |
---|
[9] | 398 | } |
---|
[48] | 399 | |
---|
[146] | 400 | |
---|
| 401 | |
---|
[185] | 402 | std::vector<float> SDMath::statistic(const CountedPtr<SDMemTable>& in, |
---|
| 403 | const std::vector<bool>& mask, |
---|
| 404 | const String& which) |
---|
[130] | 405 | // |
---|
| 406 | // Perhaps iteration over pol/beam/if should be in here |
---|
| 407 | // and inside the nrow iteration ? |
---|
| 408 | // |
---|
| 409 | { |
---|
| 410 | const uInt nRow = in->nRow(); |
---|
| 411 | std::vector<float> result(nRow); |
---|
| 412 | Vector<Bool> msk(mask); |
---|
| 413 | |
---|
| 414 | // Specify cursor location |
---|
| 415 | |
---|
[152] | 416 | IPosition start, end; |
---|
[185] | 417 | getCursorLocation(start, end, *in); |
---|
[130] | 418 | |
---|
| 419 | // Loop over rows |
---|
| 420 | |
---|
| 421 | const uInt nEl = msk.nelements(); |
---|
| 422 | for (uInt ii=0; ii < in->nRow(); ++ii) { |
---|
| 423 | |
---|
| 424 | // Get row and deconstruct |
---|
| 425 | |
---|
| 426 | MaskedArray<Float> marr(in->rowAsMaskedArray(ii)); |
---|
| 427 | Array<Float> arr = marr.getArray(); |
---|
| 428 | Array<Bool> barr = marr.getMask(); |
---|
| 429 | |
---|
| 430 | // Access desired piece of data |
---|
| 431 | |
---|
| 432 | Array<Float> v((arr(start,end)).nonDegenerate()); |
---|
| 433 | Array<Bool> m((barr(start,end)).nonDegenerate()); |
---|
| 434 | |
---|
| 435 | // Apply OTF mask |
---|
| 436 | |
---|
| 437 | MaskedArray<Float> tmp; |
---|
| 438 | if (m.nelements()==nEl) { |
---|
| 439 | tmp.setData(v,m&&msk); |
---|
| 440 | } else { |
---|
| 441 | tmp.setData(v,m); |
---|
| 442 | } |
---|
| 443 | |
---|
| 444 | // Get statistic |
---|
| 445 | |
---|
[144] | 446 | result[ii] = mathutil::statistics(which, tmp); |
---|
[130] | 447 | } |
---|
| 448 | // |
---|
| 449 | return result; |
---|
| 450 | } |
---|
| 451 | |
---|
[146] | 452 | |
---|
[185] | 453 | SDMemTable* SDMath::bin(const SDMemTable& in, Int width) |
---|
[144] | 454 | { |
---|
[169] | 455 | SDHeader sh = in.getSDHeader(); |
---|
| 456 | SDMemTable* pTabOut = new SDMemTable(in, True); |
---|
[163] | 457 | |
---|
[169] | 458 | // Bin up SpectralCoordinates |
---|
[163] | 459 | |
---|
[169] | 460 | IPosition factors(1); |
---|
| 461 | factors(0) = width; |
---|
| 462 | for (uInt j=0; j<in.nCoordinates(); ++j) { |
---|
| 463 | CoordinateSystem cSys; |
---|
| 464 | cSys.addCoordinate(in.getCoordinate(j)); |
---|
| 465 | CoordinateSystem cSysBin = |
---|
[185] | 466 | CoordinateUtil::makeBinnedCoordinateSystem(factors, cSys, False); |
---|
[169] | 467 | // |
---|
| 468 | SpectralCoordinate sCBin = cSysBin.spectralCoordinate(0); |
---|
| 469 | pTabOut->setCoordinate(sCBin, j); |
---|
| 470 | } |
---|
[163] | 471 | |
---|
[169] | 472 | // Use RebinLattice to find shape |
---|
[130] | 473 | |
---|
[169] | 474 | IPosition shapeIn(1,sh.nchan); |
---|
[185] | 475 | IPosition shapeOut = RebinLattice<Float>::rebinShape(shapeIn, factors); |
---|
[169] | 476 | sh.nchan = shapeOut(0); |
---|
| 477 | pTabOut->putSDHeader(sh); |
---|
[144] | 478 | |
---|
| 479 | |
---|
[169] | 480 | // Loop over rows and bin along channel axis |
---|
| 481 | |
---|
[209] | 482 | const uInt axis = asap::ChanAxis; |
---|
[169] | 483 | for (uInt i=0; i < in.nRow(); ++i) { |
---|
| 484 | SDContainer sc = in.getSDContainer(i); |
---|
[144] | 485 | // |
---|
[169] | 486 | Array<Float> tSys(sc.getTsys()); // Get it out before sc changes shape |
---|
[144] | 487 | |
---|
[169] | 488 | // Bin up spectrum |
---|
[144] | 489 | |
---|
[169] | 490 | MaskedArray<Float> marr(in.rowAsMaskedArray(i)); |
---|
| 491 | MaskedArray<Float> marrout; |
---|
| 492 | LatticeUtilities::bin(marrout, marr, axis, width); |
---|
[144] | 493 | |
---|
[169] | 494 | // Put back the binned data and flags |
---|
[144] | 495 | |
---|
[169] | 496 | IPosition ip2 = marrout.shape(); |
---|
| 497 | sc.resize(ip2); |
---|
[146] | 498 | // |
---|
[185] | 499 | putDataInSDC(sc, marrout.getArray(), marrout.getMask()); |
---|
[146] | 500 | |
---|
[169] | 501 | // Bin up Tsys. |
---|
[146] | 502 | |
---|
[169] | 503 | Array<Bool> allGood(tSys.shape(),True); |
---|
| 504 | MaskedArray<Float> tSysIn(tSys, allGood, True); |
---|
[146] | 505 | // |
---|
[169] | 506 | MaskedArray<Float> tSysOut; |
---|
| 507 | LatticeUtilities::bin(tSysOut, tSysIn, axis, width); |
---|
| 508 | sc.putTsys(tSysOut.getArray()); |
---|
[146] | 509 | // |
---|
[169] | 510 | pTabOut->putSDContainer(sc); |
---|
| 511 | } |
---|
| 512 | return pTabOut; |
---|
[146] | 513 | } |
---|
| 514 | |
---|
[185] | 515 | SDMemTable* SDMath::simpleOperate(const SDMemTable& in, Float val, Bool doAll, |
---|
| 516 | uInt what) |
---|
[152] | 517 | // |
---|
| 518 | // what = 0 Multiply |
---|
| 519 | // 1 Add |
---|
[146] | 520 | { |
---|
[152] | 521 | SDMemTable* pOut = new SDMemTable(in,False); |
---|
| 522 | const Table& tOut = pOut->table(); |
---|
| 523 | ArrayColumn<Float> spec(tOut,"SPECTRA"); |
---|
[146] | 524 | // |
---|
[152] | 525 | if (doAll) { |
---|
| 526 | for (uInt i=0; i < tOut.nrow(); i++) { |
---|
| 527 | |
---|
| 528 | // Get |
---|
| 529 | |
---|
| 530 | MaskedArray<Float> marr(pOut->rowAsMaskedArray(i)); |
---|
| 531 | |
---|
| 532 | // Operate |
---|
| 533 | |
---|
| 534 | if (what==0) { |
---|
| 535 | marr *= val; |
---|
| 536 | } else if (what==1) { |
---|
| 537 | marr += val; |
---|
| 538 | } |
---|
| 539 | |
---|
| 540 | // Put |
---|
| 541 | |
---|
| 542 | spec.put(i, marr.getArray()); |
---|
| 543 | } |
---|
| 544 | } else { |
---|
| 545 | |
---|
| 546 | // Get cursor location |
---|
| 547 | |
---|
| 548 | IPosition start, end; |
---|
[185] | 549 | getCursorLocation(start, end, in); |
---|
[152] | 550 | // |
---|
| 551 | for (uInt i=0; i < tOut.nrow(); i++) { |
---|
| 552 | |
---|
| 553 | // Get |
---|
| 554 | |
---|
| 555 | MaskedArray<Float> dataIn(pOut->rowAsMaskedArray(i)); |
---|
| 556 | |
---|
| 557 | // Modify. More work than we would like to deal with the mask |
---|
| 558 | |
---|
| 559 | Array<Float>& values = dataIn.getRWArray(); |
---|
| 560 | Array<Bool> mask(dataIn.getMask()); |
---|
| 561 | // |
---|
| 562 | Array<Float> values2 = values(start,end); |
---|
| 563 | Array<Bool> mask2 = mask(start,end); |
---|
| 564 | MaskedArray<Float> t(values2,mask2); |
---|
| 565 | if (what==0) { |
---|
| 566 | t *= val; |
---|
| 567 | } else if (what==1) { |
---|
| 568 | t += val; |
---|
| 569 | } |
---|
| 570 | values(start, end) = t.getArray(); // Write back into 'dataIn' |
---|
| 571 | |
---|
| 572 | // Put |
---|
| 573 | spec.put(i, dataIn.getArray()); |
---|
| 574 | } |
---|
| 575 | } |
---|
| 576 | // |
---|
[146] | 577 | return pOut; |
---|
| 578 | } |
---|
| 579 | |
---|
| 580 | |
---|
[152] | 581 | |
---|
[185] | 582 | SDMemTable* SDMath::averagePol(const SDMemTable& in, const Vector<Bool>& mask) |
---|
[152] | 583 | // |
---|
[165] | 584 | // Average all polarizations together, weighted by variance |
---|
| 585 | // |
---|
| 586 | { |
---|
| 587 | // WeightType wtType = NONE; |
---|
[185] | 588 | // convertWeightString(wtType, weight); |
---|
[165] | 589 | |
---|
| 590 | const uInt nRows = in.nRow(); |
---|
[209] | 591 | const uInt polAxis = asap::PolAxis; // Polarization axis |
---|
| 592 | const uInt chanAxis = asap::ChanAxis; // Spectrum axis |
---|
[165] | 593 | |
---|
| 594 | // Create output Table and reshape number of polarizations |
---|
| 595 | |
---|
| 596 | Bool clear=True; |
---|
| 597 | SDMemTable* pTabOut = new SDMemTable(in, clear); |
---|
| 598 | SDHeader header = pTabOut->getSDHeader(); |
---|
| 599 | header.npol = 1; |
---|
| 600 | pTabOut->putSDHeader(header); |
---|
| 601 | |
---|
| 602 | // Shape of input and output data |
---|
| 603 | |
---|
| 604 | const IPosition& shapeIn = in.rowAsMaskedArray(0u, False).shape(); |
---|
| 605 | IPosition shapeOut(shapeIn); |
---|
| 606 | shapeOut(polAxis) = 1; // Average all polarizations |
---|
| 607 | // |
---|
| 608 | const uInt nChan = shapeIn(chanAxis); |
---|
| 609 | const IPosition vecShapeOut(4,1,1,1,nChan); // A multi-dim form of a Vector shape |
---|
| 610 | IPosition start(4), end(4); |
---|
| 611 | |
---|
| 612 | // Output arrays |
---|
| 613 | |
---|
| 614 | Array<Float> outData(shapeOut, 0.0); |
---|
| 615 | Array<Bool> outMask(shapeOut, True); |
---|
| 616 | const IPosition axes(2, 2, 3); // pol-channel plane |
---|
| 617 | // |
---|
| 618 | const Bool useMask = (mask.nelements() == shapeIn(chanAxis)); |
---|
| 619 | |
---|
| 620 | // Loop over rows |
---|
| 621 | |
---|
| 622 | for (uInt iRow=0; iRow<nRows; iRow++) { |
---|
| 623 | |
---|
| 624 | // Get data for this row |
---|
| 625 | |
---|
| 626 | MaskedArray<Float> marr(in.rowAsMaskedArray(iRow)); |
---|
| 627 | Array<Float>& arr = marr.getRWArray(); |
---|
| 628 | const Array<Bool>& barr = marr.getMask(); |
---|
| 629 | |
---|
| 630 | // Make iterators to iterate by pol-channel planes |
---|
| 631 | |
---|
| 632 | ReadOnlyArrayIterator<Float> itDataPlane(arr, axes); |
---|
| 633 | ReadOnlyArrayIterator<Bool> itMaskPlane(barr, axes); |
---|
| 634 | |
---|
| 635 | // Accumulations |
---|
| 636 | |
---|
| 637 | Float fac = 1.0; |
---|
| 638 | Vector<Float> vecSum(nChan,0.0); |
---|
| 639 | |
---|
| 640 | // Iterate through data by pol-channel planes |
---|
| 641 | |
---|
| 642 | while (!itDataPlane.pastEnd()) { |
---|
| 643 | |
---|
| 644 | // Iterate through plane by polarization and accumulate Vectors |
---|
| 645 | |
---|
| 646 | Vector<Float> t1(nChan); t1 = 0.0; |
---|
| 647 | Vector<Bool> t2(nChan); t2 = True; |
---|
| 648 | MaskedArray<Float> vecSum(t1,t2); |
---|
| 649 | Float varSum = 0.0; |
---|
| 650 | { |
---|
| 651 | ReadOnlyVectorIterator<Float> itDataVec(itDataPlane.array(), 1); |
---|
| 652 | ReadOnlyVectorIterator<Bool> itMaskVec(itMaskPlane.array(), 1); |
---|
| 653 | while (!itDataVec.pastEnd()) { |
---|
| 654 | |
---|
| 655 | // Create MA of data & mask (optionally including OTF mask) and get variance |
---|
| 656 | |
---|
| 657 | if (useMask) { |
---|
| 658 | const MaskedArray<Float> spec(itDataVec.vector(),mask&&itMaskVec.vector()); |
---|
| 659 | fac = 1.0 / variance(spec); |
---|
| 660 | } else { |
---|
| 661 | const MaskedArray<Float> spec(itDataVec.vector(),itMaskVec.vector()); |
---|
| 662 | fac = 1.0 / variance(spec); |
---|
| 663 | } |
---|
| 664 | |
---|
| 665 | // Normalize spectrum (without OTF mask) and accumulate |
---|
| 666 | |
---|
| 667 | const MaskedArray<Float> spec(fac*itDataVec.vector(), itMaskVec.vector()); |
---|
| 668 | vecSum += spec; |
---|
| 669 | varSum += fac; |
---|
| 670 | |
---|
| 671 | // Next |
---|
| 672 | |
---|
| 673 | itDataVec.next(); |
---|
| 674 | itMaskVec.next(); |
---|
| 675 | } |
---|
| 676 | } |
---|
| 677 | |
---|
| 678 | // Normalize summed spectrum |
---|
| 679 | |
---|
| 680 | vecSum /= varSum; |
---|
| 681 | |
---|
| 682 | // FInd position in input data array. We are iterating by pol-channel |
---|
| 683 | // plane so all that will change is beam and IF and that's what we want. |
---|
| 684 | |
---|
| 685 | IPosition pos = itDataPlane.pos(); |
---|
| 686 | |
---|
| 687 | // Write out data. This is a bit messy. We have to reform the Vector |
---|
| 688 | // accumulator into an Array of shape (1,1,1,nChan) |
---|
| 689 | |
---|
| 690 | start = pos; |
---|
| 691 | end = pos; |
---|
| 692 | end(chanAxis) = nChan-1; |
---|
| 693 | outData(start,end) = vecSum.getArray().reform(vecShapeOut); |
---|
| 694 | outMask(start,end) = vecSum.getMask().reform(vecShapeOut); |
---|
| 695 | |
---|
| 696 | // Step to next beam/IF combination |
---|
| 697 | |
---|
| 698 | itDataPlane.next(); |
---|
| 699 | itMaskPlane.next(); |
---|
| 700 | } |
---|
| 701 | |
---|
| 702 | // Generate output container and write it to output table |
---|
| 703 | |
---|
| 704 | SDContainer sc = in.getSDContainer(); |
---|
| 705 | sc.resize(shapeOut); |
---|
| 706 | // |
---|
[185] | 707 | putDataInSDC(sc, outData, outMask); |
---|
[165] | 708 | pTabOut->putSDContainer(sc); |
---|
| 709 | } |
---|
| 710 | // |
---|
| 711 | return pTabOut; |
---|
| 712 | } |
---|
[167] | 713 | |
---|
[169] | 714 | |
---|
[185] | 715 | SDMemTable* SDMath::smooth(const SDMemTable& in, |
---|
| 716 | const casa::String& kernelType, |
---|
| 717 | casa::Float width, Bool doAll) |
---|
[177] | 718 | { |
---|
[169] | 719 | |
---|
[177] | 720 | // Number of channels |
---|
[169] | 721 | |
---|
[209] | 722 | const uInt chanAxis = asap::ChanAxis; // Spectral axis |
---|
[177] | 723 | SDHeader sh = in.getSDHeader(); |
---|
| 724 | const uInt nChan = sh.nchan; |
---|
| 725 | |
---|
| 726 | // Generate Kernel |
---|
| 727 | |
---|
[185] | 728 | VectorKernel::KernelTypes type = VectorKernel::toKernelType(kernelType); |
---|
[177] | 729 | Vector<Float> kernel = VectorKernel::make(type, width, nChan, True, False); |
---|
| 730 | |
---|
| 731 | // Generate Convolver |
---|
| 732 | |
---|
| 733 | IPosition shape(1,nChan); |
---|
| 734 | Convolver<Float> conv(kernel, shape); |
---|
| 735 | |
---|
| 736 | // New Table |
---|
| 737 | |
---|
| 738 | SDMemTable* pTabOut = new SDMemTable(in,True); |
---|
| 739 | |
---|
| 740 | // Get cursor location |
---|
| 741 | |
---|
| 742 | IPosition start, end; |
---|
[185] | 743 | getCursorLocation(start, end, in); |
---|
[177] | 744 | // |
---|
| 745 | IPosition shapeOut(4,1); |
---|
| 746 | |
---|
| 747 | // Output Vectors |
---|
| 748 | |
---|
| 749 | Vector<Float> valuesOut(nChan); |
---|
| 750 | Vector<Bool> maskOut(nChan); |
---|
| 751 | |
---|
| 752 | // Loop over rows in Table |
---|
| 753 | |
---|
| 754 | for (uInt ri=0; ri < in.nRow(); ++ri) { |
---|
| 755 | |
---|
| 756 | // Get copy of data |
---|
| 757 | |
---|
| 758 | const MaskedArray<Float>& dataIn(in.rowAsMaskedArray(ri)); |
---|
| 759 | AlwaysAssert(dataIn.shape()(chanAxis)==nChan, AipsError); |
---|
| 760 | // |
---|
| 761 | Array<Float> valuesIn = dataIn.getArray(); |
---|
| 762 | Array<Bool> maskIn = dataIn.getMask(); |
---|
| 763 | |
---|
| 764 | // Branch depending on whether we smooth all locations or just |
---|
| 765 | // those pointed at by the current selection cursor |
---|
| 766 | |
---|
| 767 | if (doAll) { |
---|
| 768 | uInt axis = 3; |
---|
| 769 | VectorIterator<Float> itValues(valuesIn, axis); |
---|
| 770 | VectorIterator<Bool> itMask(maskIn, axis); |
---|
| 771 | while (!itValues.pastEnd()) { |
---|
| 772 | |
---|
| 773 | // Smooth |
---|
| 774 | if (kernelType==VectorKernel::HANNING) { |
---|
| 775 | mathutil::hanning(valuesOut, maskOut, itValues.vector(), itMask.vector()); |
---|
| 776 | itMask.vector() = maskOut; |
---|
| 777 | } else { |
---|
| 778 | mathutil::replaceMaskByZero(itValues.vector(), itMask.vector()); |
---|
| 779 | conv.linearConv(valuesOut, itValues.vector()); |
---|
| 780 | } |
---|
| 781 | // |
---|
| 782 | itValues.vector() = valuesOut; |
---|
| 783 | // |
---|
| 784 | itValues.next(); |
---|
| 785 | itMask.next(); |
---|
| 786 | } |
---|
| 787 | } else { |
---|
| 788 | |
---|
| 789 | // Set multi-dim Vector shape |
---|
| 790 | |
---|
| 791 | shapeOut(chanAxis) = valuesIn.shape()(chanAxis); |
---|
| 792 | |
---|
| 793 | // Stuff about with shapes so that we don't have conformance run-time errors |
---|
| 794 | |
---|
| 795 | Vector<Float> valuesIn2 = valuesIn(start,end).nonDegenerate(); |
---|
| 796 | Vector<Bool> maskIn2 = maskIn(start,end).nonDegenerate(); |
---|
| 797 | |
---|
| 798 | // Smooth |
---|
| 799 | |
---|
| 800 | if (kernelType==VectorKernel::HANNING) { |
---|
| 801 | mathutil::hanning(valuesOut, maskOut, valuesIn2, maskIn2); |
---|
| 802 | maskIn(start,end) = maskOut.reform(shapeOut); |
---|
| 803 | } else { |
---|
| 804 | mathutil::replaceMaskByZero(valuesIn2, maskIn2); |
---|
| 805 | conv.linearConv(valuesOut, valuesIn2); |
---|
| 806 | } |
---|
| 807 | // |
---|
| 808 | valuesIn(start,end) = valuesOut.reform(shapeOut); |
---|
| 809 | } |
---|
| 810 | |
---|
| 811 | // Create and put back |
---|
| 812 | |
---|
| 813 | SDContainer sc = in.getSDContainer(ri); |
---|
[185] | 814 | putDataInSDC(sc, valuesIn, maskIn); |
---|
[177] | 815 | // |
---|
| 816 | pTabOut->putSDContainer(sc); |
---|
| 817 | } |
---|
| 818 | // |
---|
| 819 | return pTabOut; |
---|
| 820 | } |
---|
| 821 | |
---|
| 822 | |
---|
| 823 | |
---|
| 824 | |
---|
| 825 | |
---|
[169] | 826 | // 'private' functions |
---|
| 827 | |
---|
[185] | 828 | void SDMath::fillSDC(SDContainer& sc, |
---|
| 829 | const Array<Bool>& mask, |
---|
| 830 | const Array<Float>& data, |
---|
| 831 | const Array<Float>& tSys, |
---|
| 832 | Int scanID, Double timeStamp, |
---|
| 833 | Double interval, const String& sourceName, |
---|
| 834 | const Vector<uInt>& freqID) |
---|
[167] | 835 | { |
---|
[169] | 836 | // Data and mask |
---|
[167] | 837 | |
---|
[185] | 838 | putDataInSDC(sc, data, mask); |
---|
[167] | 839 | |
---|
[169] | 840 | // TSys |
---|
| 841 | |
---|
| 842 | sc.putTsys(tSys); |
---|
| 843 | |
---|
| 844 | // Time things |
---|
| 845 | |
---|
| 846 | sc.timestamp = timeStamp; |
---|
| 847 | sc.interval = interval; |
---|
| 848 | sc.scanid = scanID; |
---|
[167] | 849 | // |
---|
[169] | 850 | sc.sourcename = sourceName; |
---|
| 851 | sc.putFreqMap(freqID); |
---|
| 852 | } |
---|
[167] | 853 | |
---|
[185] | 854 | void SDMath::normalize(MaskedArray<Float>& sum, |
---|
[169] | 855 | const Array<Float>& sumSq, |
---|
| 856 | const Array<Float>& nPts, |
---|
| 857 | WeightType wtType, Int axis, |
---|
| 858 | Int nAxesSub) |
---|
| 859 | { |
---|
| 860 | IPosition pos2(nAxesSub,0); |
---|
| 861 | // |
---|
| 862 | if (wtType==NONE) { |
---|
[167] | 863 | |
---|
[169] | 864 | // We just average by the number of points accumulated. |
---|
| 865 | // We need to make a MA out of nPts so that no divide by |
---|
| 866 | // zeros occur |
---|
[167] | 867 | |
---|
[169] | 868 | MaskedArray<Float> t(nPts, (nPts>Float(0.0))); |
---|
| 869 | sum /= t; |
---|
| 870 | } else if (wtType==VAR) { |
---|
[167] | 871 | |
---|
[169] | 872 | // Normalize each spectrum by sum(1/var) where the variance |
---|
| 873 | // is worked out for each spectrum |
---|
| 874 | |
---|
| 875 | Array<Float>& data = sum.getRWArray(); |
---|
| 876 | VectorIterator<Float> itData(data, axis); |
---|
| 877 | while (!itData.pastEnd()) { |
---|
| 878 | pos2 = itData.pos().getFirst(nAxesSub); |
---|
| 879 | itData.vector() /= sumSq(pos2); |
---|
| 880 | itData.next(); |
---|
| 881 | } |
---|
| 882 | } else if (wtType==TSYS) { |
---|
| 883 | } |
---|
| 884 | } |
---|
| 885 | |
---|
| 886 | |
---|
[185] | 887 | void SDMath::accumulate(Double& timeSum, Double& intSum, Int& nAccum, |
---|
| 888 | MaskedArray<Float>& sum, Array<Float>& sumSq, |
---|
| 889 | Array<Float>& nPts, Array<Float>& tSysSum, |
---|
| 890 | const Array<Float>& tSys, const Array<Float>& nInc, |
---|
| 891 | const Vector<Bool>& mask, Double time, Double interval, |
---|
| 892 | const Block<CountedPtr<SDMemTable> >& in, |
---|
| 893 | uInt iTab, uInt iRow, uInt axis, |
---|
| 894 | uInt nAxesSub, Bool useMask, |
---|
| 895 | WeightType wtType) |
---|
[169] | 896 | { |
---|
| 897 | |
---|
| 898 | // Get data |
---|
| 899 | |
---|
| 900 | MaskedArray<Float> dataIn(in[iTab]->rowAsMaskedArray(iRow)); |
---|
| 901 | Array<Float>& valuesIn = dataIn.getRWArray(); // writable reference |
---|
| 902 | const Array<Bool>& maskIn = dataIn.getMask(); // RO reference |
---|
[167] | 903 | // |
---|
[169] | 904 | if (wtType==NONE) { |
---|
| 905 | const MaskedArray<Float> n(nInc,dataIn.getMask()); |
---|
| 906 | nPts += n; // Only accumulates where mask==T |
---|
| 907 | } else if (wtType==VAR) { |
---|
[167] | 908 | |
---|
[169] | 909 | // We are going to average the data, weighted by the noise for each pol, beam and IF. |
---|
| 910 | // So therefore we need to iterate through by spectrum (axis 3) |
---|
[167] | 911 | |
---|
[169] | 912 | VectorIterator<Float> itData(valuesIn, axis); |
---|
| 913 | ReadOnlyVectorIterator<Bool> itMask(maskIn, axis); |
---|
| 914 | Float fac = 1.0; |
---|
| 915 | IPosition pos(nAxesSub,0); |
---|
| 916 | // |
---|
| 917 | while (!itData.pastEnd()) { |
---|
[167] | 918 | |
---|
[169] | 919 | // Make MaskedArray of Vector, optionally apply OTF mask, and find scaling factor |
---|
[167] | 920 | |
---|
[169] | 921 | if (useMask) { |
---|
| 922 | MaskedArray<Float> tmp(itData.vector(),mask&&itMask.vector()); |
---|
| 923 | fac = 1.0/variance(tmp); |
---|
| 924 | } else { |
---|
| 925 | MaskedArray<Float> tmp(itData.vector(),itMask.vector()); |
---|
| 926 | fac = 1.0/variance(tmp); |
---|
| 927 | } |
---|
| 928 | |
---|
| 929 | // Scale data |
---|
| 930 | |
---|
| 931 | itData.vector() *= fac; // Writes back into 'dataIn' |
---|
[167] | 932 | // |
---|
[169] | 933 | // Accumulate variance per if/pol/beam averaged over spectrum |
---|
| 934 | // This method to get pos2 from itData.pos() is only valid |
---|
| 935 | // because the spectral axis is the last one (so we can just |
---|
| 936 | // copy the first nAXesSub positions out) |
---|
[167] | 937 | |
---|
[169] | 938 | pos = itData.pos().getFirst(nAxesSub); |
---|
| 939 | sumSq(pos) += fac; |
---|
| 940 | // |
---|
| 941 | itData.next(); |
---|
| 942 | itMask.next(); |
---|
| 943 | } |
---|
| 944 | } else if (wtType==TSYS) { |
---|
| 945 | } |
---|
[167] | 946 | |
---|
[169] | 947 | // Accumulate sum of (possibly scaled) data |
---|
| 948 | |
---|
| 949 | sum += dataIn; |
---|
| 950 | |
---|
| 951 | // Accumulate Tsys, time, and interval |
---|
| 952 | |
---|
| 953 | tSysSum += tSys; |
---|
| 954 | timeSum += time; |
---|
| 955 | intSum += interval; |
---|
| 956 | nAccum += 1; |
---|
| 957 | } |
---|
| 958 | |
---|
| 959 | |
---|
| 960 | |
---|
| 961 | |
---|
[185] | 962 | void SDMath::getCursorLocation(IPosition& start, IPosition& end, |
---|
| 963 | const SDMemTable& in) |
---|
[169] | 964 | { |
---|
| 965 | const uInt nDim = 4; |
---|
| 966 | const uInt i = in.getBeam(); |
---|
| 967 | const uInt j = in.getIF(); |
---|
| 968 | const uInt k = in.getPol(); |
---|
| 969 | const uInt n = in.nChan(); |
---|
[167] | 970 | // |
---|
[169] | 971 | start.resize(nDim); |
---|
| 972 | start(0) = i; |
---|
| 973 | start(1) = j; |
---|
| 974 | start(2) = k; |
---|
| 975 | start(3) = 0; |
---|
[167] | 976 | // |
---|
[169] | 977 | end.resize(nDim); |
---|
| 978 | end(0) = i; |
---|
| 979 | end(1) = j; |
---|
| 980 | end(2) = k; |
---|
| 981 | end(3) = n-1; |
---|
| 982 | } |
---|
| 983 | |
---|
| 984 | |
---|
[185] | 985 | void SDMath::convertWeightString(WeightType& wtType, const std::string& weightStr) |
---|
[169] | 986 | { |
---|
| 987 | String tStr(weightStr); |
---|
| 988 | tStr.upcase(); |
---|
| 989 | if (tStr.contains(String("NONE"))) { |
---|
| 990 | wtType = NONE; |
---|
| 991 | } else if (tStr.contains(String("VAR"))) { |
---|
| 992 | wtType = VAR; |
---|
| 993 | } else if (tStr.contains(String("TSYS"))) { |
---|
| 994 | wtType = TSYS; |
---|
[185] | 995 | throw(AipsError("T_sys weighting not yet implemented")); |
---|
[169] | 996 | } else { |
---|
[185] | 997 | throw(AipsError("Unrecognized weighting type")); |
---|
[167] | 998 | } |
---|
| 999 | } |
---|
| 1000 | |
---|
[185] | 1001 | void SDMath::putDataInSDC(SDContainer& sc, const Array<Float>& data, |
---|
| 1002 | const Array<Bool>& mask) |
---|
[169] | 1003 | { |
---|
| 1004 | sc.putSpectrum(data); |
---|
| 1005 | // |
---|
| 1006 | Array<uChar> outflags(data.shape()); |
---|
| 1007 | convertArray(outflags,!mask); |
---|
| 1008 | sc.putFlags(outflags); |
---|
| 1009 | } |
---|