def test(self, queryData, vectorNormalize, vectorDifference, output):
file = h5py.File(output, 'w')
results = file.create_dataset('matches', (queryData.size, self.dataset.size), dtype=numpy.float64, compression='gzip', compression_opts=9)
if queryData == self.dataset:
# If self-testing, do not match an image with itself
utility.iterate(self.representations, lambda representation, i: results.__setitem__(i, self.matchRepresentationOthers(representation, i, vectorNormalize, vectorDifference)), False)
else:
# Else, test every query image against the complete set
utility.iterate(queryData.images(), lambda image, i: results.__setitem__(i, self.matchRepresentation(self.representImage(image), vectorNormalize, vectorDifference)), False)
file.close()
def zip(dx, dy):
result = [[], []]
for key, x in sorted(dx.iteritems()):
for y in utility.iterate(dy.get(key)):
result[0].append(x)
result[1].append(y)
return result
def generateScale(scale, note, mode=1, rType="list", octaves=True): #scale, start, type
"""
Generate a scale
scale (string): major, melodic_minor, harmonic_minor, chromatic, major_pentatonic
note: start note
"""
if scale in SCALE_STEPS:
steps = _get_mode(SCALE_STEPS[scale], mode)
return iterate(note, steps, rType, octaves)
else:
raise MelopyGenericError("Unknown scale type:" + str(scale))
def generateScale(scale, note, mode=1, rType="list", octaves=True): #scale, start, type
"""
Generate a scale
scale (string): major, melodic_minor, harmonic_minor, chromatic, major_pentatonic
note: start note
"""
if scale in SCALE_STEPS:
steps = _get_mode(SCALE_STEPS[scale], mode)
return iterate(note, steps, rType, octaves)
else:
raise MelopyGenericError("Unknown scale type:" + str(scale))
def imageDifferences(matrix, image): # Return a sorted list of the differences between one image and the others in a given matrix if len(matrix) == len(matrix[0]): # Matrix is square, assume self-comparison differences = numpy.empty((len(matrix)-1, 2), dtype=numpy.float64) row = matrix[image] i = 0 while i < image: # All similarities before the diagonal differences[i] = numpy.array([row[i], i]) i += 1 while i < differences.shape[0]: # All similarities beyond the diagonal differences[i] = numpy.array([row[i + 1], i + 1]) i += 1 else: # Non-square matrix, must be differing source and query differences = numpy.empty((len(matrix[0]), 2), dtype=numpy.float64) utility.iterate(matrix[image], lambda difference, i: differences.__setitem__(i, (difference, i)), False, None) return differences[differences[:,0].argsort()]
def generateChord(name, tonic, inversion=0, rType='list', octaves=True):
if name in CHORD_INTERVALS:
steps = CHORD_INTERVALS[name]
return _get_inversion(iterate(tonic, steps, rType, octaves),inversion)
else:
raise MelopyGenericError("Unknown Chord:"+str(name))
def test_iterate(self):
start = 'D4'
pattern = [2, 2, 1, 2, 2, 2]
should_be = ['D4', 'E4', 'F#4', 'G4', 'A4', 'B4', 'C#5']
self.assertEqual(should_be, utility.iterate(start, pattern))
def fileMetrics(result, sourceData, queryData, amounts):
similarityMatrix = utility.duration(
readSimilarities,
sourceData.name,
queryData.name,
task="Read similarity matrix"
)
differenceFile, differenceMatrix = utility.duration(
readDifferences,
result,
task="Read difference matrix"
)
sourcePoses = utility.duration(
readPoses,
sourceData.name,
task="Read poses",
)
queryPoses = utility.duration(
readPoses,
queryData.name,
task="Read poses",
)
totalMetrics = {
'precision': {amount:numpy.empty(queryData.size, dtype=numpy.float64) for amount in amounts},
'recall': {amount:numpy.empty(queryData.size, dtype=numpy.float64) for amount in amounts},
'recallRate': {amount:numpy.empty(queryData.size, dtype=numpy.float64) for amount in amounts},
'transform': numpy.empty(queryData.size, dtype=numpy.float64),
'rotation': numpy.empty(queryData.size, dtype=numpy.float64)
}
utility.iterate(range(queryData.size), lambda _, i: registerMetrics(totalMetrics, imageMetrics(similarityMatrix, differenceMatrix, sourcePoses, queryPoses, i, amounts), i), False)
differenceFile.close()
return {
'precision': {
amount: {
'mean': numpy.mean(totalMetrics['precision'][amount]),
'variance': numpy.var(totalMetrics['precision'][amount]),
'standardDeviation': numpy.std(totalMetrics['precision'][amount])
} for amount in totalMetrics['precision']
},
'recall': {
amount: {
'mean': numpy.mean(totalMetrics['recall'][amount]),
'variance': numpy.var(totalMetrics['recall'][amount]),
'standardDeviation': numpy.std(totalMetrics['recall'][amount])
} for amount in totalMetrics['recall']
},
'recallRate': {
amount: {
'mean': numpy.mean(totalMetrics['recallRate'][amount]),
'variance': numpy.var(totalMetrics['recallRate'][amount]),
'standardDeviation': numpy.std(totalMetrics['recallRate'][amount])
} for amount in totalMetrics['recallRate']
},
'transform': {
'mean': numpy.mean(totalMetrics['transform']),
'variance': numpy.var(totalMetrics['transform']),
'standardDeviation': numpy.std(totalMetrics['transform'])
},
'rotation': {
'mean': numpy.mean(totalMetrics['rotation']),
'variance': numpy.var(totalMetrics['rotation']),
'standardDeviation': numpy.std(totalMetrics['rotation'])
}
}
def matchRepresentation(self, query, vectorNormalize, vectorDifference): # Returns a list of the differences between the given representation and each representation in the dataset matches = numpy.empty(self.dataset.size, dtype=numpy.float64) utility.iterate(self.representations, lambda representation, i: (matches.__setitem__(i, self.compareRepresentation(query, representation, vectorNormalize, vectorDifference))), False, None) return matches
def representDataset(self): representations = numpy.empty((self.dataset.size, self.representationSize()), dtype=numpy.int16) utility.iterate(self.dataset.images(), lambda image, i: (representations.__setitem__(i, self.representImage(image))), False) return representations