rh4SrSSKJr SSKJr SSKJr Sr"SS5rS S jr S r \ S :XaSSKr\R"5 gg) z This module takes two XML files and displays the number of measures that differ between the two before and after running the combined correction models ) annotations) correctors) converterFc\rSrSrSrSSjr\S5r\RS5rSr Sr S r S r S r S rS rSrSrSrg)OmrGroundTruthPairz Object for making comparisons between an OMR score and the GroundTruth Takes in a path to the OMR and a path to the groundTruth (or a pair of music21.stream.Score objects). See below for examples. NcLSUlSUl[US5(aURUlXlO XlSUlUR 5Ul[US5(aURUlX l O X lSUl UR5Ul g)NfilePath) _overriddenDebugnumberOfDifferenceshasattrr omrPath omrM21Score getOmrScoreomrScore groundPathgroundM21ScoregetGroundScore groundScore)selfomrgrounds P/home/james-whalen/.local/lib/python3.13/site-packages/music21/omr/evaluators.py__init__OmrGroundTruthPair.__init__)s $#' 3 # #<s"  ( (( (rcXlg)N)r )rnewDebugs rr r!Is (rcXUR5UlUR5Ulg)z Parse both scores. N)rrrrrs rparseAllOmrGroundTruthPair.parseAllMs%((* ..0rclURR5 URR5 g)z# store the Hashes for both scores. N)r getAllHashesrrs rhashAllOmrGroundTruthPair.hashAllTs& ""$ %%'rcUR(a [S5 UR(d%[R"UR 5Ul[ R"UR5$)af Returns a ScoreCorrector object of the OMR score. does not store it anywhere >>> omrPath = omr.correctors.K525omrShortPath >>> ground = omr.correctors.K525groundTruthShortPath >>> omrGTP = omr.evaluators.OmrGroundTruthPair(omr=omrPath, ground=ground) >>> ssOMR = omrGTP.getOmrScore() >>> ssOMR zparsing OMR score)r printrrparserrScoreCorrectorrs rrOmrGroundTruthPair.getOmrScore[sI :: % &(t||>> omrPath = omr.correctors.K525omrShortPath >>> ground = omr.correctors.K525groundTruthShortPath >>> omrGTP = omr.evaluators.OmrGroundTruthPair(omr=omrPath, ground=ground) >>> ssGT = omrGTP.getGroundScore() >>> ssGT zparsing Ground Truth score)r r,rrr-rrr.rs rr!OmrGroundTruthPair.getGroundScorensI :: . /"""+//$//"BD (()<)<==rcX:Xagg)zL define the substitution cost for x and y (2 if x and y are unequal else 0) r)rxys r substCostOmrGroundTruthPair.substCosts 6rcg)z+ define the insertion cost for x and y (1) r4rr5s r insertCostOmrGroundTruthPair.insertCostrcg)z* define the deletion cost for x and y (1) r:r4r;s r deleteCostOmrGroundTruthPair.deleteCostr>rc [U5n[U5n[US-5VVs/sH"n[US-5Vs/sHnSPM snPM$ nnn[SUS-5H)nXvS- SURXS- 5-XvS'M+ [SUS-5H+nUSUS- URX%S- 5-USU'M- [SUS-5Hin[SUS-5HSn[ XvS- US-XvUS- S-XvS- US- UR X%S- XS- 5-5XvU'MU Mk XsU$s snfs snnf)zC Computes the min edit distance from target to source. Figure 3.25 r:r)lenranger<r@minr7)rtargetsourcenmjidistances r minEditDistOmrGroundTruthPair.minEditDistst K K7>> omrPath = omr.correctors.K525omrShortPath >>> ground = omr.correctors.K525groundTruthShortPath >>> omrGTP = omr.evaluators.OmrGroundTruthPair(omr=omrPath, ground=ground) >>> differences = omrGTP.getDifferences() >>> differences 32 r)r rr(rrDrCrM)romrListgtListpartNumomrPartgtPart measureErrorss rgetDifferences!OmrGroundTruthPair.getDifferencess$% --,,.!!..0S\*G&G_F ,,W=M  $ $ 5 $ + '''r)r rrrr rrr)NN)__name__ __module__ __qualname____firstlineno____doc__rpropertyr setterr%r)rrr7r<r@rMrV__static_attributes__r4rrrrsf1*)) \\))1(;&>J  .(rrNc(Uc[nSn[XS9nU(a [S5 UcUR5nOUnU(a [SU5 URnUn U(a [S5 /n Sn Sn USLafU R Un U R 5nU(a/[SU5 [S U R UR5 U R5 O[[U R 55HvnU R Un U R 5nU [U5- n U R5nU RU5 U [U R UR5- n Mx U(a8[S 5 [S 5 [S U 5 [S 5 [S5 U R5nU(a8[S 5 [S 5 [SU5 [S 5 [S5 U RU U5nU(a[S5 [U5 UR5nU(a>[SU5 [SU 5 [SU 5 U RR5 UUU U S.nU$)aK Get a dictionary showing the efficacy of the omr.correctors.ScoreCorrector on an OMR Score by comparing it to the GroundTruth. Set debug to True to see a lot of intermediary steps. >>> omrFilePath = omr.correctors.K525omrShortPath >>> groundTruthFilePath = omr.correctors.K525groundTruthShortPath >>> returnDict = omr.evaluators.evaluateCorrectingModel(omrFilePath, groundTruthFilePath) >>> for name in sorted(list(returnDict.keys())): ... (name, returnDict[name]) ('newEditDistance', 20) ('numberOfFlaggedMeasures', 13) ('originalEditDistance', 32) ('totalNumberOfMeasures', 84) r:)rrzgetting differenceszOriginal edit distancez2Running Horizontal Model (Prior-based-on-distance)rTzIncorrect measure indices:z Hashed notes:z+for each entry in the array below, we have zt[flagged measure part, flagged measure index, source measure part, source measure index, source measure probability]zHORIZONTAL CORRECTING ARRAYz"**********************************z-Running Vertical Model (Prior-based-on-Parts)zVERTICAL CORRECTING MEASURESz]Finding best from Horizontal and Vertical and replacing flagged measures with source measuresz4done replacing flagged measures with source measuresznew edit distancez%number of flagged measures originallyztotal number of measures)originalEditDistancenewEditDistancenumberOfFlaggedMeasurestotalNumberOfMeasures)rrr,rVr singlePartsgetIncorrectMeasureIndices hashedNotesrunHorizontalCorrectionModelrDrCappendrunVerticalCorrectionModelgenerateCorrectedScorescoreshow)rgroundTruthPathr originalDifferences runOnePartpnomrGTPr myOmrScorescorrectingArrayHorAllPartnumberOfIncorrectMeasuresnumberOfTotalMeasures scorePartincorrectMeasureIndicestempPNcorrectingArrayHorOnePartcorrectingArrayVertAllPart priorScorenewNumberOfDifferences returnDicts revaluateCorrectingModelrsk& } B G DF  #$"$3351  &(;<JA  BC " !TMM"% "+"F"F"H  .0G H /1==#4#@#@ A..0C ./F f-I&/&J&J&L # %-D)E E %(1(N(N(P % % , ,-F G !Sv)>)J)J%K K ! 0  ;< D E +-FG 23 =>!"!=!=!?  ;< E F ,.HI 23 0 1))*CE_`J  DE j#224  !#9: 57PQ (*?@  *=%;-F+@BJ rc[R"U5nUR5nSnSnURn[ U5HupgXVR 5n[ U5HzupX;aM US- nSn [ U5HupX:XaM X:XdMSn O U SLa2[ [U55HnX:XaM X.U nX:XdMSn O U SLdMuUS- nM| M X44$)aW Essentially it's the ratio of amount of rhythmic similarity within a piece, which gives an upper bound on what the omr.corrector.prior measure should be able to achieve for the flagged measures. If a piece has low rhythmic similarity in general, then there's no way for a correct match to be found within the unflagged measures in the piece. Returns a tuple of the total number of NON-flagged measures and the total number of those measures that have a rhythmic match. Takes in a stream.Score. >>> c = converter.parse(omr.correctors.K525omrShortPath) # first 21 measures >>> totalUnflagged, totalUnflaggedWithMatches = omr.evaluators.autoCorrelationBestMeasure(c) >>> (totalUnflagged, totalUnflaggedWithMatches) (71, 64) >>> print( float(totalUnflaggedWithMatches) / totalUnflagged ) 0.901... Schoenberg has low autoCorrelation. >>> c = corpus.parse('schoenberg/opus19/movement6') >>> totalUnflagged, totalUnflaggedWithMatches = omr.evaluators.autoCorrelationBestMeasure(c) >>> (totalUnflagged, totalUnflaggedWithMatches) (18, 6) >>> print( float(totalUnflaggedWithMatches) / totalUnflagged ) 0.333... rr:FT)rr.r(re enumeraterfrDrC) inputScoress allHashes totalMeasures totalMatchesrepNum pHashArrayincorrectIndicesrKmHashmatchrJnHash otherPNum otherHashs rautoCorrelationBestMeasurer>s<  " ": .B!IML..K%i0&,GGI!*-HA$ Q ME&j16> E 2~!&s;'7!8I ( ) 4Q 7I ) $ "9}! 5.1:  ((r__main__)NNF) r\ __future__r music21.omrrmusic21rrrrrrXmainTestr4rrrs^#" w(w(t=AAFcLC)L z r