def annotationsFi2instances(annFi,
m,
tf,
labelsHierarchy=None,
gaps='b',
sep=None):
"""reads annotations from file into instances
Parameters
----------
annFi: str
path to file
m: int
number of instances
labelsHierarchy: list like object
hierarchy of the annotations in case of overlaps.
Only one label per instance, eg = ['c', 'w']
gaps: string,
label for the gaps
Returns
-------
labels_arr: ndarray (m, )
array with the instance labels"""
if labelsHierarchy is None:
labelsHierarchy = ['c', 'w']
T, L = annT.anns2TLndarrays(annFi, sep=sep)
if tf == 'auto':
tf = T[-1, -1] + 0.1
return annotations2instanceArray(T,
L,
m,
tf,
labelsHierarchy=labelsHierarchy)
def annotationsFi2instances(annFi, m, tf, labelsHierarchy=None, gaps="b"):
"""reads annotations from file into instances"""
if labelsHierarchy is None:
labelsHierarchy = ["c", "w"]
T, L = annT.anns2TLndarrays(annFi)
if tf == "auto":
tf = T[-1, -1] + 0.1
return annotations2instanceArray(T, L, m, tf, labelsHierarchy=labelsHierarchy)
def test_predictions():
# predict annotations
wF = os.path.join(pDir, 'data/WAV_0111_001-48kHz_30-60sec.wav')
anF = os.path.join(pDir, 'data/WAV_0111_001-48kHz_30-60sec.txt')
predsF1 = os.path.join(pDir, 'data/WAV_0111_001-48kHz_30-60sec-preds0.txt')
pre.predictAnnotationSections(wF, anF, clf, feExFun, lt, outFile=predsF1 )
predsF2 = os.path.join(pDir, 'data/WAV_0111_001-48kHz_30-60sec-preds0.txt')
pre.predictAnnotationSections0(wF, anF, clf, feExFun, lt, outFile=predsF2 )
## load annotations into T, L format
Tfi1, Lfi1 = annT.anns2TLndarrays( predsF1, Tcols=(0,1), Lcols=(-1,))
Tfi2, Lfi2 = annT.anns2TLndarrays( predsF2, Tcols=(0,1), Lcols=(-1,))
# test both prediction types
np.testing.assert_array_equal(Lfi1, Lfi2)
np.testing.assert_array_equal(Tfi1, Tfi2)
# test one more prediction mode
Tp, Lp = pre.TLpredictAnnotationSections( wF, anF, clf, feExFun, lt)
np.testing.assert_array_equal(Lfi1, Lp)
np.testing.assert_array_equal(Tfi1, Tp)
def TLpredictAnnotationSections(wavF,
annF,
clf,
featExtFun,
lt,
printProbs=False,
readSections=None,
printreadSectionsC=True):
"""generates annotations predicting audio section classes
Parameters
----------
wavF : str
annF : str
path to the file with the annotation section to predict
clf : estimator
featExtFun : callable
lt : labelTransformer
printProbs : bool
readSections : list of str
regions in the annF for which we predict
printreadSectionsC : bool
"""
## load annotations
waveform, fs = sT.wav2waveform(wavF)
T, L0 = annT.anns2TLndarrays(annF)
## set of annotation-sections to predict
if readSections is None:
readSections = np.array(list(set(L0)))
## filter for sections of interest
IO_sections = np.isin(L0, readSections)
Tp = T[IO_sections]
L = L0[IO_sections]
Lp = np.zeros_like(L)
## for each annotation section
for i, label in enumerate(L): # for each section
waveformSec = auf.getWavSec(waveform, fs,
*Tp[i]) # load waveform section
M0 = featExtFun(waveformSec) # extract features
M = np.expand_dims(M0.flatten(), axis=0)
Lp[i] = lt.num2nom(clf.predict(M))[0] # predict
return Tp, Lp
def __init__(self, annFi):
self.file_path = annFi
self.T, self.L = annT.anns2TLndarrays(self.file_path)
self.tf = self.T[-1, -1]
def predictAnnotationSections(wavF,
annF,
clf,
featExtFun,
lt,
outFile=None,
sep='\t',
printProbs=False,
header='',
readSections=None,
printreadSectionsC=True):
"""predicts annotations for call types sections
Parameters
----------
wavF: str
annF: str
clf: estimator
featExtFun: callable
lt: labelTransformer
outFil: str
sep: str
printProbs: bool
header: str
readSections: list of str
regions in the annF for which we predict
printreadSectionsC: bool
See also
--------
TLpredictAnnotationSections
TODO: recode to use TLpredictAnnotationSections
"""
if outFile is None:
outFile = os.path.splitext(annF)[0] + '-sectionPredictions.txt'
try: # remove file if exists
os.remove(outFile)
except OSError:
pass
## load files
waveform, fs = sT.wav2waveform(wavF)
T, L = annT.anns2TLndarrays(annF)
if readSections == None:
readSections = list(set(L))
## for each annotation section
for i, label in enumerate(L):
if label in readSections:
waveformSec = auf.getWavSec(waveform, fs, *T[i])
## predict
try:
M0 = featExtFun(waveformSec) # estract features
M = np.expand_dims(M0.flatten(), axis=0)
y_pred = lt.num2nom(clf.predict(M)) # predict label
except AssertionError:
y_pred = [label]
## write
with open(outFile, 'a') as f:
f.write("{}\t{}\t{}\t{}\n".format(T[i, 0], T[i, 1], label,
*y_pred))
elif printreadSectionsC:
with open(outFile, 'a') as f:
f.write("{}\t{}\t{}\t{}\n".format(T[i, 0], T[i, 1], label,
label))
return outFile