def test_missing(self):
fm = FeatureMatrix('test',self.missing_info)
self.assertEqual(fm.features, ['feature1','feature2'])
self.assertEqual(fm.possible_values,{'+','-'})
#Error, there should be a default default value?
fm.validate()
def test_missing(self):
fm = FeatureMatrix('test', self.missing_info)
self.assertEqual(fm.features, ['feature1', 'feature2'])
self.assertEqual(fm.possible_values, {'+', '-'})
#Error, there should be a default default value?
fm.validate()
def test_add_segment(self):
fm = FeatureMatrix('test',self.basic_info)
fm.add_segment('e',{'feature1':'+','feature2':'-'})
self.assertEqual(fm['e','feature1'],'+')
#Fails, need to raise exception if the added segment contains a feature that no other segment has
self.assertRaises(AttributeError,fm.add_segment,'e',{'feature1':'+','feature3':'-'})
def load_feature_matrix_csv(name, path, delimiter, stop_check = None, call_back = None):
"""
Load a FeatureMatrix from a column-delimited text file
Parameters
----------
name : str
Informative identifier to refer to feature system
path : str
Full path to text file
delimiter : str
Character to use for spliting lines into columns
stop_check : callable, optional
Optional function to check whether to gracefully terminate early
call_back : callable, optional
Optional function to supply progress information during the function
Returns
-------
FeatureMatrix
FeatureMatrix generated from the text file
"""
text_input = []
with open(path, encoding='utf-8-sig', mode='r') as f:
reader = DictReader(f, delimiter = delimiter)
lines = list(reader)
if call_back is not None:
call_back('Reading file...')
call_back(0, len(lines))
for i, line in enumerate(lines):
if stop_check is not None and stop_check():
return
if call_back is not None:
call_back(i)
if line:
if len(line.keys()) == 1:
raise(DelimiterError)
if 'symbol' not in line:
raise(KeyError)
#Compat
newline = {}
for k,v in line.items():
if k == 'symbol':
newline[k] = v
elif v is not None:
newline[k] = v[0]
text_input.append(newline)
feature_matrix = FeatureMatrix(name,text_input)
feature_matrix.validate()
return feature_matrix
def load_feature_matrix_csv(name, path, delimiter, stop_check = None, call_back = None):
"""
Load a FeatureMatrix from a column-delimited text file
Parameters
----------
name : str
Informative identifier to refer to feature system
path : str
Full path to text file
delimiter : str
Character to use for spliting lines into columns
stop_check : callable, optional
Optional function to check whether to gracefully terminate early
call_back : callable, optional
Optional function to supply progress information during the function
Returns
-------
FeatureMatrix
FeatureMatrix generated from the text file
"""
text_input = []
with open(path, encoding='utf-8-sig', mode='r') as f:
reader = DictReader(f, delimiter = delimiter)
lines = list(reader)
if call_back is not None:
call_back('Reading file...')
call_back(0, len(lines))
for i, line in enumerate(lines):
if stop_check is not None and stop_check():
return
if call_back is not None:
call_back(i)
if line:
if len(line.keys()) == 1:
raise(DelimiterError)
if 'symbol' not in line:
raise(KeyError)
#Compat
newline = {}
for k,v in line.items():
if k == 'symbol':
newline[k] = v
elif v is not None:
newline[k] = v[0]
text_input.append(newline)
feature_matrix = FeatureMatrix(name,text_input)
feature_matrix.validate()
return feature_matrix
def test_missing_with_default(self):
fm = FeatureMatrix('test',self.missing_with_default_info)
self.assertEqual(fm.features, ['feature1','feature2'])
self.assertEqual(fm.possible_values,{'+','-','n'})
fm.validate()
self.assertEqual(fm['b','feature2'], 'n')
def test_missing_with_default(self):
fm = FeatureMatrix('test', self.missing_with_default_info)
self.assertEqual(fm.features, ['feature1', 'feature2'])
self.assertEqual(fm.possible_values, {'+', '-', 'n'})
fm.validate()
self.assertEqual(fm['b', 'feature2'], 'n')
def test_add_segment(self):
fm = FeatureMatrix('test', self.basic_info)
fm.add_segment('e', {'feature1': '+', 'feature2': '-'})
self.assertEqual(fm['e', 'feature1'], '+')
#Fails, need to raise exception if the added segment contains a feature that no other segment has
self.assertRaises(AttributeError, fm.add_segment, 'e', {
'feature1': '+',
'feature3': '-'
})
def test_basic(self):
corpus = Corpus('test')
for w in self.corpus_basic_info:
corpus.add_word(Word(**w))
fm = FeatureMatrix('test', self.feature_basic_info)
corpus.set_feature_matrix(fm)
def test_add_feature(self):
fm = FeatureMatrix('test', self.missing_with_default_info)
fm.add_feature('feature3')
self.assertEqual(fm['a', 'feature3'], 'n')
fm = FeatureMatrix('test', self.basic_info)
#Error, no default value
fm.add_feature('feature3')
def test_coverage(self):
corpus = Corpus('test')
for w in self.corpus_basic_info:
corpus.add_word(Word(**w))
fm = FeatureMatrix('test', self.feature_no_d_info)
corpus.set_feature_matrix(fm)
self.assertEqual(corpus.check_coverage(), ['d'])
def test_basic(self):
fm = FeatureMatrix('test', self.basic_info)
self.assertTrue(fm.name == 'test')
self.assertEqual(fm.features, ['feature1', 'feature2'])
self.assertEqual(fm.possible_values, {'+', '-'})
#fails, should be sorted list?
self.assertEqual(sorted(fm.segments), sorted(['#', 'a', 'b', 'c',
'd']))
def test_dots(self):
fm = FeatureMatrix('test', self.dots_info)
self.assertEqual(fm.features, ['feature1', 'feature2'])
self.assertEqual(fm.possible_values, {'+', '-', '.'})
self.assertEqual(fm['b', 'feature2'], '.')
#Fails, should be sorted list of _features? Or set of _features? Would need to be hashed then
self.assertEqual(fm['b']['feature2'], '.')
def test_feats_to_segs(self):
corpus = Corpus('test')
for w in self.corpus_basic_info:
corpus.add_word(Word(**w))
fm = FeatureMatrix('test', self.feature_basic_info)
corpus.set_feature_matrix(fm)
self.assertEqual(sorted(corpus.features_to_segments(['+feature1'])),
sorted(['a', 'b']))
def setUp(self):
self.corpus_info = [
{
'spelling': 'a',
'transcription': ['a', 'b'],
'frequency': 32.0
},
{
'spelling': 'b',
'transcription': ['a', 'b'],
'frequency': 32.0
},
{
'spelling': 'c',
'transcription': ['c', 'a', 'b'],
'frequency': 32.0
},
{
'spelling': 'd',
'transcription': ['a', 'd'],
'frequency': 32.0
},
]
self.feature_info = [{
'symbol': 'a',
'feature1': '+',
'feature2': '+'
}, {
'symbol': 'b',
'feature1': '+',
'feature2': '-'
}, {
'symbol': 'c',
'feature1': '-',
'feature2': '+'
}, {
'symbol': 'd',
'feature1': '-',
'feature2': '-'
}]
self.corpus = Corpus('test')
for w in self.corpus_info:
self.corpus.add_word(Word(**w))
fm = FeatureMatrix('test', self.feature_info)
self.corpus.set_feature_matrix(fm)
self.corpus.inventory.update_features(self.corpus.specifier)
def test_add_tier(self):
corpus = Corpus('test')
for w in self.corpus_basic_info:
corpus.add_word(Word(**w))
fm = FeatureMatrix('test', self.feature_basic_info)
corpus.set_feature_matrix(fm)
corpus.add_tier('t', '+feature1')
self.assertEqual(corpus['d'].t, [corpus['d'].transcription[0]])
corpus.remove_attribute('t')
self.assertRaises(AttributeError, getattr, corpus['d'], 't')
def test_add_feature(self):
fm = FeatureMatrix('test',self.missing_with_default_info)
fm.add_feature('feature3')
self.assertEqual(fm['a','feature3'], 'n')
fm = FeatureMatrix('test',self.basic_info)
#Error, no default value
fm.add_feature('feature3')
def spe_specifier():
fm_input = [{'symbol':'ɑ','EXTRA':'-','LONG':'-','ant':'-','back':'+','cont':'+','cor':'-',
'del_rel':'n','distr':'n','glot_cl':'-','hi_subgl_pr':'-','high':'-',
'lat':'n','low':'+','mv_glot_cl':'n','nasal':'-','round':'-','son':'+',
'strid':'-','tense':'-','voc':'+','voice':'+'},
{'symbol':'u','EXTRA':'-','LONG':'-','ant':'-','back':'+','cont':'+','cor':'-',
'del_rel':'n','distr':'n','glot_cl':'-','hi_subgl_pr':'-','high':'+',
'lat':'n','low':'-','mv_glot_cl':'n','nasal':'-','round':'+','son':'+',
'strid':'-','tense':'+','voc':'+','voice':'+'},
{'symbol':'o','EXTRA':'-','LONG':'-','ant':'-','back':'+','cont':'+','cor':'-',
'del_rel':'n','distr':'n','glot_cl':'-','hi_subgl_pr':'-','high':'-',
'lat':'n','low':'-','mv_glot_cl':'n','nasal':'-','round':'+','son':'+',
'strid':'-','tense':'+','voc':'+','voice':'+'},
{'symbol':'e','EXTRA':'-','LONG':'-','ant':'-','back':'-','cont':'+','cor':'-',
'del_rel':'n','distr':'n','glot_cl':'-','hi_subgl_pr':'-','high':'-',
'lat':'n','low':'-','mv_glot_cl':'n','nasal':'-','round':'-','son':'+',
'strid':'-','tense':'+','voc':'+','voice':'+'},
{'symbol':'s','EXTRA':'-','LONG':'-','ant':'+','back':'-','cont':'+','cor':'+',
'del_rel':'n','distr':'-','glot_cl':'-','hi_subgl_pr':'-','high':'-',
'lat':'-','low':'-','mv_glot_cl':'n','nasal':'-','round':'-','son':'-',
'strid':'+','tense':'.','voc':'-','voice':'-'},
{'symbol':'m','EXTRA':'-','LONG':'-','ant':'+','back':'-','cont':'-','cor':'-',
'del_rel':'-','distr':'+','glot_cl':'-','hi_subgl_pr':'-','high':'-',
'lat':'n','low':'-','mv_glot_cl':'n','nasal':'+','round':'-','son':'+',
'strid':'-','tense':'-','voc':'-','voice':'+'},
{'symbol':'i','EXTRA':'-','LONG':'-','ant':'-','back':'-','cont':'+','cor':'-',
'del_rel':'n','distr':'n','glot_cl':'-','hi_subgl_pr':'-','high':'+',
'lat':'n','low':'-','mv_glot_cl':'n','nasal':'-','round':'-','son':'+',
'strid':'-','tense':'+','voc':'+','voice':'+'},
{'symbol':'n','EXTRA':'-','LONG':'-','ant':'+','back':'-','cont':'-','cor':'+',
'del_rel':'-','distr':'-','glot_cl':'-','hi_subgl_pr':'-','high':'-',
'lat':'-','low':'-','mv_glot_cl':'n','nasal':'+','round':'-','son':'+',
'strid':'-','tense':'-','voc':'-','voice':'+'},
{'symbol':'ʃ','EXTRA':'-','LONG':'-','ant':'-','back':'-','cont':'+','cor':'+',
'del_rel':'n','distr':'+','glot_cl':'-','hi_subgl_pr':'-','high':'+',
'lat':'-','low':'-','mv_glot_cl':'n','nasal':'-','round':'-','son':'-',
'strid':'+','tense':'.','voc':'-','voice':'-'},
{'symbol':'t','EXTRA':'-','LONG':'-','ant':'+','back':'-','cont':'-','cor':'+',
'del_rel':'-','distr':'-','glot_cl':'-','hi_subgl_pr':'-','high':'-',
'lat':'-','low':'-','mv_glot_cl':'n','nasal':'-','round':'-','son':'-',
'strid':'-','tense':'.','voc':'-','voice':'-'}]
fm = FeatureMatrix('spe',fm_input)
return fm
