def test_prod_wordtokens_type(specified_discourse_corpus):
seg1 = 's'
seg2 = 'ʃ'
expected = {
"-voc": 0.0,
"+voc,+high": 0.863120568566631,
"+voc,-high": 0.9852281360342515,
"#": 0.0
}
env_list = []
expected_envs = {}
for k, v in expected.items():
if k != '#':
segs = specified_discourse_corpus.lexicon.features_to_segments(k)
else:
segs = k
env = EnvironmentFilter(['s', 'ʃ'], None, [segs])
env_list.append(env)
expected_envs[env] = v
expected_envs["AVG"] = 0.9241743523004413
type_or_token = 'type'
tier = 'transcription'
with MostFrequentVariantContext(specified_discourse_corpus.lexicon, tier,
type_or_token) as c:
result = calc_prod(c, env_list, all_info=False)
for k, v in result.items():
assert (expected_envs[k] - v < 0.001)
with WeightedVariantContext(specified_discourse_corpus.lexicon, tier,
type_or_token) as c:
result = calc_prod(c, env_list, all_info=False)
for k, v in result.items():
assert (expected_envs[k] - v < 0.001)
def test_prod_wordtokens_type(specified_discourse_corpus):
seg1 = 's'
seg2 = 'ʃ'
expected = {"-voc":0.0,
"+voc,+high":0.863120568566631,
"+voc,-high":0.9852281360342515,
"#":0.0}
env_list = []
expected_envs = {}
for k, v in expected.items():
if k != '#':
segs = specified_discourse_corpus.lexicon.features_to_segments(k)
else:
segs = k
env = EnvironmentFilter(['s', 'ʃ'], None, [segs])
env_list.append(env)
expected_envs[env] = v
expected_envs["AVG"] = 0.9241743523004413
type_or_token = 'type'
tier = 'transcription'
with MostFrequentVariantContext(specified_discourse_corpus.lexicon, tier, type_or_token) as c:
result = calc_prod(c, env_list, all_info=False)
for k,v in result.items():
assert(expected_envs[k]-v < 0.001)
with WeightedVariantContext(specified_discourse_corpus.lexicon, tier, type_or_token) as c:
result = calc_prod(c, env_list, all_info=False)
for k,v in result.items():
assert(expected_envs[k]-v < 0.001)
def test_prod_wordtokens_token(specified_discourse_corpus):
seg1 = 's'
seg2 = 'ʃ'
expected = {"-voc":0.0,
"+voc,+high":0.8631205, #0.9321115676166747, #Error!!!?!?!?
"+voc,-high":0.9660096062568557,
"#":0.0}
env_list = []
expected_envs = {}
for k, v in expected.items():
if k != '#':
segs = specified_discourse_corpus.lexicon.features_to_segments(k)
else:
segs = k
env = EnvironmentFilter(['s', 'ʃ'], None, [segs])
env_list.append(env)
expected_envs[env] = v
expected_envs["AVG"] = 0.9241743523004413
type_or_token = 'token'
tier = 'transcription'
with MostFrequentVariantContext(specified_discourse_corpus.lexicon, tier, type_or_token) as c:
result = calc_prod(c, env_list)
for k,v in result.items():
assert(expected_envs[k]-v < 0.001)
type_or_token = 'token'
tier = 'transcription'
with WeightedVariantContext(specified_discourse_corpus.lexicon, tier, type_or_token) as c:
result = calc_prod(c, env_list)
for k,v in result.items():
assert(expected_envs[k]-v < 0.001)
def test_prod_token(specified_test_corpus):
seg1 = 's'
seg2 = 'ʃ'
expected = {
"-voc": 0.0,
"+voc,+high": 0.9321115676166747,
"+voc,-high": 0.9660096062568557,
"#": 0.0
}
env_list = []
expected_envs = {}
for k, v in expected.items():
if k != '#':
segs = specified_test_corpus.features_to_segments(k)
else:
segs = k
env = EnvironmentFilter(['s', 'ʃ'], None, [segs])
env_list.append(env)
expected_envs[env] = v
expected_envs["AVG"] = 0.9241743523004413
type_or_token = 'token'
tier = 'transcription'
with CanonicalVariantContext(specified_test_corpus, tier,
type_or_token) as c:
result = calc_prod(c, env_list)
for k, v in result.items():
assert (expected_envs[k] - v < 0.001)
def run(self):
kwargs = self.kwargs
self.results = []
context = kwargs.pop('context')
if context == ContextWidget.canonical_value:
cm = CanonicalVariantContext
elif context == ContextWidget.frequent_value:
cm = MostFrequentVariantContext
elif context == ContextWidget.separate_value:
cm = SeparatedTokensVariantContext
elif context == ContextWidget.relative_value:
cm = WeightedVariantContext
with cm(kwargs['corpus'],
kwargs['sequence_type'],
kwargs['type_token'],
frequency_threshold=kwargs['frequency_cutoff']) as c:
try:
envs = kwargs.pop('envs', None)
for pair in kwargs['segment_pairs']:
ordered_pair = pair
if envs is not None:
for env in envs:
env.middle = set(pair)
res = calc_prod(c,
envs,
kwargs['strict'],
ordered_pair=ordered_pair,
all_info=True,
stop_check=kwargs['stop_check'],
call_back=kwargs['call_back'])
else:
res = calc_prod_all_envs(
c,
pair[0],
pair[1],
all_info=True,
stop_check=kwargs['stop_check'],
call_back=kwargs['call_back'])
if self.stopped:
break
self.results.append(res)
except PCTError as e:
self.errorEncountered.emit(e)
return
except Exception as e:
e = PCTPythonError(e)
self.errorEncountered.emit(e)
return
if self.stopped:
self.finishedCancelling.emit()
return
self.dataReady.emit(self.results)
def run(self):
kwargs = self.kwargs
self.results = []
context = kwargs.pop('context')
if context == ContextWidget.canonical_value:
cm = CanonicalVariantContext
elif context == ContextWidget.frequent_value:
cm = MostFrequentVariantContext
elif context == ContextWidget.separate_value:
cm = SeparatedTokensVariantContext
elif context == ContextWidget.relative_value:
cm = WeightedVariantContext
with cm(kwargs['corpus'], kwargs['sequence_type'], kwargs['type_token'],
frequency_threshold = kwargs['frequency_cutoff']) as c:
try:
envs = kwargs.pop('envs', None)
for pair in kwargs['segment_pairs']:
ordered_pair = pair
if envs is not None:
for env in envs:
env.middle = set(pair)
res = calc_prod(c,
envs,
kwargs['strict'],
ordered_pair = ordered_pair,
all_info = True,
stop_check = kwargs['stop_check'],
call_back = kwargs['call_back'])
else:
res = calc_prod_all_envs(c, pair[0], pair[1],
all_info = True,
stop_check = kwargs['stop_check'],
call_back = kwargs['call_back'])
if self.stopped:
break
self.results.append(res)
except PCTError as e:
self.errorEncountered.emit(e)
return
except Exception as e:
e = PCTPythonError(e)
self.errorEncountered.emit(e)
return
if self.stopped:
self.finishedCancelling.emit()
return
self.dataReady.emit(self.results)
