def test_should_compute_tokenization_shift(self):
# Given
text = " hello? world"
tokens = tokenize(text, "en")
# When
shifts = _compute_char_shifts(tokens)
# Then
expected_shifts = [-2, -2, -2, -2, -2, -1, -1, -3, -3, -3, -3, -3, -3]
self.assertListEqual(expected_shifts, shifts)
def _parse(self, text, scope):
tokens = tokenize(text, self.language)
shifts = _compute_char_shifts(tokens)
cleaned_text = " ".join(token.value for token in tokens)
entities = self._parser.parse(cleaned_text, scope)
for entity in entities:
start = entity[RES_MATCH_RANGE][START]
end = entity[RES_MATCH_RANGE][END]
entity[ENTITY_KIND] = entity.pop("entity_identifier")
entity[RES_MATCH_RANGE][START] -= shifts[start]
entity[RES_MATCH_RANGE][END] -= shifts[end - 1]
return entities
def test_entity_match_factory(self):
# Given
config = {
"factory_name": "entity_match",
"args": {
"tagging_scheme_code": TaggingScheme.BILOU.value,
"use_stemming": True
},
"offsets": [0]
}
tokens = tokenize("2 dummy a had dummy_c", LANGUAGE_EN)
cache = [{TOKEN_NAME: token} for token in tokens]
factory = get_feature_factory(config)
dataset = deepcopy(SAMPLE_DATASET)
dataset = validate_and_format_dataset(dataset)
custom_entity_parser = CustomEntityParser.build(
dataset, CustomEntityParserUsage.WITH_STEMS)
factory.fit(dataset, "dummy_intent_1")
# When
features = factory.build_features(
custom_entity_parser=custom_entity_parser)
features = sorted(features, key=lambda f: f.base_name)
res0 = features[0].compute(0, cache)
res1 = features[0].compute(1, cache)
res2 = features[0].compute(2, cache)
res3 = features[0].compute(3, cache)
res4 = features[0].compute(4, cache)
res5 = features[1].compute(0, cache)
res6 = features[1].compute(1, cache)
res7 = features[1].compute(2, cache)
res8 = features[1].compute(3, cache)
res9 = features[1].compute(4, cache)
# Then
self.assertIsInstance(factory, CustomEntityMatchFactory)
self.assertEqual(len(features), 2)
self.assertEqual(features[0].base_name, "entity_match_dummy_entity_1")
self.assertEqual(features[1].base_name, "entity_match_dummy_entity_2")
self.assertEqual(res0, BEGINNING_PREFIX)
self.assertEqual(res1, INSIDE_PREFIX)
self.assertEqual(res2, LAST_PREFIX)
self.assertEqual(res3, None)
self.assertEqual(res4, None)
self.assertEqual(res5, None)
self.assertEqual(res6, None)
self.assertEqual(res7, None)
self.assertEqual(res8, None)
self.assertEqual(res9, UNIT_PREFIX)
def test_custom_multi_feature_factory(self):
# Given
# pylint:disable=unused-variable
@CRFFeatureFactory.register("my_multi_feature_factory", override=True)
class MyMultiFeature(CRFFeatureFactory):
def build_features(self):
first_features = [
Feature("my_first_feature",
self.compute_feature_1,
offset=offset) for offset in self.offsets
]
second_features = [
Feature("my_second_feature",
self.compute_feature_2,
offset=offset) for offset in self.offsets
]
return first_features + second_features
@staticmethod
def compute_feature_1(tokens, token_index):
return "(%s)[my_feature_1]" % tokens[token_index].value
@staticmethod
def compute_feature_2(tokens, token_index):
return "(%s)[my_feature_2]" % tokens[token_index].value
# pylint:enable=unused-variable
# When
config = {
"factory_name": "my_multi_feature_factory",
"args": {},
"offsets": [-1, 0]
}
feature_factory = CRFFeatureFactory.from_config(config)
features = feature_factory.build_features()
feature_0 = features[0]
feature_1 = features[1]
feature_2 = features[2]
feature_3 = features[3]
tokens = tokenize("foo bar baz", "en")
cache = [{TOKEN_NAME: token} for token in tokens]
# Then
self.assertEqual("my_first_feature[-1]", feature_0.name)
self.assertEqual("(foo)[my_feature_1]", feature_0.compute(1, cache))
self.assertEqual("my_first_feature", feature_1.name)
self.assertEqual("my_second_feature[-1]", feature_2.name)
self.assertEqual("(bar)[my_feature_2]", feature_2.compute(2, cache))
self.assertEqual("my_second_feature", feature_3.name)
def get_slots(self, text):
tokens = tokenize(text, self.language)
slots = []
for token in tokens:
normalized_value = token.value
if self.config.get("lowercase", False):
normalized_value = normalized_value.lower()
if normalized_value in self.slots_keywords:
entity = self.slots_keywords[normalized_value][0]
slot_name = self.slots_keywords[normalized_value][1]
slot = unresolved_slot((token.start, token.end), token.value,
entity, slot_name)
slots.append(slot)
return slots
def test_should_tokenize_literals(self):
# Given
language = LANGUAGE_EN
text = "Hello Beautiful World"
# When
tokens = tokenize(text, language)
# Then
expected_tokens = [
Token(value='Hello', start=0, end=5),
Token(value='Beautiful', start=6, end=15),
Token(value='World', start=16, end=21)
]
self.assertListEqual(tokens, expected_tokens)
def test_should_tokenize_symbols(self):
# Given
language = LANGUAGE_EN
text = "$$ % !!"
# When
tokens = tokenize(text, language)
# Then
expected_tokens = [
Token(value='$$', start=0, end=2),
Token(value='%', start=3, end=4),
Token(value='!!', start=5, end=7)
]
self.assertListEqual(tokens, expected_tokens)
def test_length_factory(self):
# Given
config = {"factory_name": "length", "args": {}, "offsets": [0]}
tokens = tokenize("hello beautiful world", LANGUAGE_EN)
cache = [{TOKEN_NAME: token} for token in tokens]
factory = CRFFeatureFactory.from_config(config)
factory.fit(None, None)
features = factory.build_features()
# When
res = features[0].compute(2, cache)
# Then
self.assertIsInstance(factory, LengthFactory)
self.assertEqual(features[0].base_name, "length")
self.assertEqual(res, "5")
def test_feature_should_work_with_offset(self):
# Given
def fn(tokens, token_index):
value = tokens[token_index].value
return "%s_%s" % (value, len(value))
cache = [{
TOKEN_NAME: token
} for token in tokenize("hello beautiful world", LANGUAGE_EN)]
feature = Feature("test_feature", fn, offset=1)
# When
res = feature.compute(1, cache)
# Then
self.assertEqual(res, "world_5")
def _preprocess_text(self, string):
"""Replaces stop words and characters that are tokenized out by
whitespaces"""
tokens = tokenize(string, self.language)
current_idx = 0
cleaned_string = ""
for token in tokens:
if self.stop_words and normalize_token(token) in self.stop_words:
token.value = "".join(" " for _ in range(len(token.value)))
prefix_length = token.start - current_idx
cleaned_string += "".join((" " for _ in range(prefix_length)))
cleaned_string += token.value
current_idx = token.end
suffix_length = len(string) - current_idx
cleaned_string += "".join((" " for _ in range(suffix_length)))
return cleaned_string
def test_should_compute_features(self):
# Given
features_factories = [
{
"factory_name": NgramFactory.name,
"args": {
"n": 1,
"use_stemming": False,
"common_words_gazetteer_name": None
},
"offsets": [0],
"drop_out": 0.3
},
]
slot_filler_config = CRFSlotFillerConfig(
feature_factory_configs=features_factories, random_seed=40)
tokens = tokenize("foo hello world bar", LANGUAGE_EN)
dataset_stream = io.StringIO("""
---
type: intent
name: my_intent
utterances:
- this is [slot1:entity1](my first entity)
- this is [slot2:entity2](second_entity)""")
dataset = Dataset.from_yaml_files("en", [dataset_stream]).json
shared = self.get_shared_data(dataset,
CustomEntityParserUsage.WITHOUT_STEMS)
slot_filler = CRFSlotFiller(slot_filler_config, **shared)
slot_filler.fit(dataset, intent="my_intent")
# When
features_with_drop_out = slot_filler.compute_features(tokens, True)
# Then
expected_features = [
{
"ngram_1": "foo"
},
{},
{
"ngram_1": "world"
},
{},
]
self.assertListEqual(expected_features, features_with_drop_out)
def test_is_first_factory(self):
# Given
config = {"factory_name": "is_first", "args": {}, "offsets": [0]}
tokens = tokenize("hello beautiful world", LANGUAGE_EN)
cache = [{TOKEN_NAME: token} for token in tokens]
factory = CRFFeatureFactory.from_config(config)
factory.fit(None, None)
features = factory.build_features()
# When
res1 = features[0].compute(0, cache)
res2 = features[0].compute(1, cache)
# Then
self.assertIsInstance(factory, IsFirstFactory)
self.assertEqual(features[0].base_name, "is_first")
self.assertEqual(res1, "1")
self.assertEqual(res2, None)
def _parse(self, text, scope=None):
tokens = tokenize(text, self.language)
shifts = _compute_char_shifts(tokens)
cleaned_text = " ".join(token.value for token in tokens)
entities = self._parser.parse(cleaned_text, scope)
result = []
for entity in entities:
start = entity["range"]["start"]
start -= shifts[start]
end = entity["range"]["end"]
end -= shifts[end - 1]
entity_range = {START: start, END: end}
ent = parsed_entity(entity_kind=entity["entity_identifier"],
entity_value=entity["value"],
entity_resolved_value=entity["resolved_value"],
entity_range=entity_range)
result.append(ent)
return result
def get_slots(self, text):
"""Extracts slots from the provided text
Returns:
list of dict: The list of extracted slots
Raises:
NotTrained: When the slot filler is not fitted
"""
if not self.slot_name_mapping:
# Early return if the intent has no slots
return []
tokens = tokenize(text, self.language)
if not tokens:
return []
features = self.compute_features(tokens)
tags = [_decode_tag(tag) for tag in
self.crf_model.predict_single(features)]
return tags_to_slots(text, tokens, tags, self.config.tagging_scheme,
self.slot_name_mapping)
def test_suffix_factory(self):
# Given
config = {
"factory_name": "suffix",
"args": {
"suffix_size": 2
},
"offsets": [0]
}
tokens = tokenize("hello beautiful world", LANGUAGE_EN)
cache = [{TOKEN_NAME: token} for token in tokens]
factory = get_feature_factory(config)
factory.fit(None, None)
features = factory.build_features()
# When
res = features[0].compute(1, cache)
# Then
self.assertIsInstance(factory, SuffixFactory)
self.assertEqual(features[0].base_name, "suffix_2")
self.assertEqual(res, "ul")
def test_word_cluster_factory(self):
# Given
resources = {
WORD_CLUSTERS: {
"my_word_clusters": {
"word1": "00",
"word2": "11"
}
}
}
config = {
"factory_name": "word_cluster",
"args": {
"cluster_name": "my_word_clusters",
"use_stemming": False
},
"offsets": [0]
}
tokens = tokenize("hello word1 word2", LANGUAGE_EN)
cache = [{TOKEN_NAME: token} for token in tokens]
factory = CRFFeatureFactory.from_config(config, resources=resources)
mocked_dataset = {"language": "en"}
factory.fit(mocked_dataset, None)
features = factory.build_features()
# When
res0 = features[0].compute(0, cache)
res1 = features[0].compute(1, cache)
res2 = features[0].compute(2, cache)
# Then
self.assertIsInstance(factory, WordClusterFactory)
self.assertEqual(features[0].base_name,
"word_cluster_my_word_clusters")
self.assertEqual(res0, None)
self.assertEqual(res1, "00")
self.assertEqual(res2, "11")
def test_filter_overlapping_builtins(self):
# Given
language = LANGUAGE_EN
text = "Find me a flight before 10pm and after 8pm"
tokens = tokenize(text, language)
tags = ['O' for _ in range(5)] + ['B-flight'] + ['O' for _ in range(3)]
tagging_scheme = TaggingScheme.BIO
builtin_entities = [{
RES_MATCH_RANGE: {
START: 17,
END: 28
},
VALUE: "before 10pm",
ENTITY_KIND: SNIPS_DATETIME
}, {
RES_MATCH_RANGE: {
START: 33,
END: 42
},
VALUE: "after 8pm",
ENTITY_KIND: SNIPS_DATETIME
}]
# When
entities = _filter_overlapping_builtins(builtin_entities, tokens, tags,
tagging_scheme)
# Then
expected_entities = [{
RES_MATCH_RANGE: {
START: 33,
END: 42
},
VALUE: "after 8pm",
ENTITY_KIND: SNIPS_DATETIME
}]
self.assertEqual(entities, expected_entities)
def test_should_compute_features(self):
# Given
features_factories = [
{
"factory_name": NgramFactory.name,
"args": {
"n": 1,
"use_stemming": False,
"common_words_gazetteer_name": None
},
"offsets": [0],
"drop_out": 0.3
},
]
slot_filler_config = CRFSlotFillerConfig(
feature_factory_configs=features_factories, random_seed=40)
slot_filler = CRFSlotFiller(slot_filler_config)
tokens = tokenize("foo hello world bar", LANGUAGE_EN)
dataset = validate_and_format_dataset(SAMPLE_DATASET)
slot_filler.fit(dataset, intent="dummy_intent_1")
# When
features_with_drop_out = slot_filler.compute_features(tokens, True)
# Then
expected_features = [
{
"ngram_1": "foo"
},
{},
{
"ngram_1": "world"
},
{},
]
self.assertListEqual(expected_features, features_with_drop_out)
def _replace_tokenized_out_characters(string, language, replacement_char=" "):
"""Replace all characters that are tokenized out by `replacement_char`
Examples:
>>> string = "hello, it's me"
>>> language = "en"
>>> tokenize_light(string, language)
['hello', 'it', 's', 'me']
>>> _replace_tokenized_out_characters(string, language, "_")
'hello__it_s_me'
"""
tokens = tokenize(string, language)
current_idx = 0
cleaned_string = ""
for token in tokens:
prefix_length = token.start - current_idx
cleaned_string += "".join(
(replacement_char for _ in range(prefix_length)))
cleaned_string += token.value
current_idx = token.end
suffix_length = len(string) - current_idx
cleaned_string += "".join((replacement_char for _ in range(suffix_length)))
return cleaned_string
def test_word_cluster_factory(self, mock_get_word_clusters):
# Given
def mocked_get_word_clusters(language):
if language == LANGUAGE_EN:
return {"mocked_cluster": {"word1": "00", "word2": "11"}}
return dict()
mock_get_word_clusters.side_effect = mocked_get_word_clusters
config = {
"factory_name": "word_cluster",
"args": {
"cluster_name": "mocked_cluster",
"use_stemming": False
},
"offsets": [0]
}
tokens = tokenize("hello word1 word2", LANGUAGE_EN)
cache = [{TOKEN_NAME: token} for token in tokens]
factory = get_feature_factory(config)
mocked_dataset = {"language": "en"}
factory.fit(mocked_dataset, None)
features = factory.build_features()
# When
res0 = features[0].compute(0, cache)
res1 = features[0].compute(1, cache)
res2 = features[0].compute(2, cache)
# Then
self.assertIsInstance(factory, WordClusterFactory)
self.assertEqual(features[0].base_name, "word_cluster_mocked_cluster")
self.assertEqual(res0, None)
self.assertEqual(res1, "00")
self.assertEqual(res2, "11")
def test_shape_ngram_factory(self):
# Given
config = {
"factory_name": "shape_ngram",
"args": {
"n": 3,
},
"offsets": [0]
}
tokens = tokenize("hello Beautiful foObar world", LANGUAGE_EN)
cache = [{TOKEN_NAME: token} for token in tokens]
factory = CRFFeatureFactory.from_config(config)
mocked_dataset = {"language": "en"}
factory.fit(mocked_dataset, None)
features = factory.build_features()
# When
res = features[0].compute(1, cache)
# Then
self.assertIsInstance(factory, ShapeNgramFactory)
self.assertEqual(features[0].base_name, "shape_ngram_3")
self.assertEqual(res, "Xxx xX xxx")
def sort_key_fn(slot):
tokens = tokenize(slot[RES_VALUE], language)
return -(len(tokens) + len(slot[RES_VALUE]))
def test_augment_slots(self):
# Given
language = LANGUAGE_EN
text = "Find me a flight before 10pm and after 8pm"
tokens = tokenize(text, language)
missing_slots = {"start_date", "end_date"}
tags = ['O' for _ in tokens]
def mocked_sequence_probability(_, tags_):
tags_1 = [
'O', 'O', 'O', 'O',
'%sstart_date' % BEGINNING_PREFIX,
'%sstart_date' % INSIDE_PREFIX, 'O',
'%send_date' % BEGINNING_PREFIX,
'%send_date' % INSIDE_PREFIX
]
tags_2 = [
'O', 'O', 'O', 'O',
'%send_date' % BEGINNING_PREFIX,
'%send_date' % INSIDE_PREFIX, 'O',
'%sstart_date' % BEGINNING_PREFIX,
'%sstart_date' % INSIDE_PREFIX
]
tags_3 = ['O', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'O']
tags_4 = [
'O', 'O', 'O', 'O', 'O', 'O', 'O',
'%sstart_date' % BEGINNING_PREFIX,
'%sstart_date' % INSIDE_PREFIX
]
tags_5 = [
'O', 'O', 'O', 'O', 'O', 'O', 'O',
'%send_date' % BEGINNING_PREFIX,
'%send_date' % INSIDE_PREFIX
]
tags_6 = [
'O', 'O', 'O', 'O',
'%sstart_date' % BEGINNING_PREFIX,
'%sstart_date' % INSIDE_PREFIX, 'O', 'O', 'O'
]
tags_7 = [
'O', 'O', 'O', 'O',
'%send_date' % BEGINNING_PREFIX,
'%send_date' % INSIDE_PREFIX, 'O', 'O', 'O'
]
tags_8 = [
'O', 'O', 'O', 'O',
'%sstart_date' % BEGINNING_PREFIX,
'%sstart_date' % INSIDE_PREFIX, 'O',
'%sstart_date' % BEGINNING_PREFIX,
'%sstart_date' % INSIDE_PREFIX
]
tags_9 = [
'O', 'O', 'O', 'O',
'%send_date' % BEGINNING_PREFIX,
'%send_date' % INSIDE_PREFIX, 'O',
'%send_date' % BEGINNING_PREFIX,
'%send_date' % INSIDE_PREFIX
]
if tags_ == tags_1:
return 0.6
elif tags_ == tags_2:
return 0.8
elif tags_ == tags_3:
return 0.2
elif tags_ == tags_4:
return 0.2
elif tags_ == tags_5:
return 0.99
elif tags_ == tags_6:
return 0.0
elif tags_ == tags_7:
return 0.0
elif tags_ == tags_8:
return 0.5
elif tags_ == tags_9:
return 0.5
else:
raise ValueError("Unexpected tag sequence: %s" % tags_)
slot_filler_config = CRFSlotFillerConfig(random_seed=42)
slot_filler = CRFSlotFiller(config=slot_filler_config)
slot_filler.language = LANGUAGE_EN
slot_filler.intent = "intent1"
slot_filler.slot_name_mapping = {
"start_date": "snips/datetime",
"end_date": "snips/datetime",
}
# pylint:disable=protected-access
slot_filler._get_sequence_probability = MagicMock(
side_effect=mocked_sequence_probability)
# pylint:enable=protected-access
slot_filler.compute_features = MagicMock(return_value=None)
# When
# pylint: disable=protected-access
augmented_slots = slot_filler._augment_slots(text, tokens, tags,
missing_slots)
# pylint: enable=protected-access
# Then
expected_slots = [
unresolved_slot(value='after 8pm',
match_range={
START: 33,
END: 42
},
entity='snips/datetime',
slot_name='end_date')
]
self.assertListEqual(augmented_slots, expected_slots)
def test_bilou_tags_to_slots(self):
# Given
language = LANGUAGE_EN
slot_name = "animal"
intent_slots_mapping = {"animal": "animal"}
tags = [
{
"text": "",
"tags": [],
"expected_slots": []
},
{
"text": "nothing here",
"tags": [OUTSIDE, OUTSIDE],
"expected_slots": []
},
{
"text":
"i am a blue bird",
"tags": [
OUTSIDE, OUTSIDE, OUTSIDE, BEGINNING_PREFIX + slot_name,
LAST_PREFIX + slot_name
],
"expected_slots": [
unresolved_slot(match_range=(7, 16),
value="blue bird",
entity=slot_name,
slot_name=slot_name)
]
},
{
"text":
"i am a bird",
"tags": [OUTSIDE, OUTSIDE, OUTSIDE, UNIT_PREFIX + slot_name],
"expected_slots": [
unresolved_slot(match_range=(7, 11),
value="bird",
entity=slot_name,
slot_name=slot_name)
]
},
{
"text":
"bird",
"tags": [UNIT_PREFIX + slot_name],
"expected_slots": [
unresolved_slot(match_range=(0, 4),
value="bird",
entity=slot_name,
slot_name=slot_name)
]
},
{
"text":
"blue bird",
"tags":
[BEGINNING_PREFIX + slot_name, LAST_PREFIX + slot_name],
"expected_slots": [
unresolved_slot(match_range=(0, 9),
value="blue bird",
entity=slot_name,
slot_name=slot_name)
]
},
{
"text":
"light blue bird blue bird",
"tags": [
BEGINNING_PREFIX + slot_name, INSIDE_PREFIX + slot_name,
LAST_PREFIX + slot_name, BEGINNING_PREFIX + slot_name,
LAST_PREFIX + slot_name
],
"expected_slots": [
unresolved_slot(match_range=(0, 15),
value="light blue bird",
entity=slot_name,
slot_name=slot_name),
unresolved_slot(match_range=(16, 25),
value="blue bird",
entity=slot_name,
slot_name=slot_name)
]
},
{
"text":
"bird birdy",
"tags": [UNIT_PREFIX + slot_name, UNIT_PREFIX + slot_name],
"expected_slots": [
unresolved_slot(match_range=(0, 4),
value="bird",
entity=slot_name,
slot_name=slot_name),
unresolved_slot(match_range=(5, 10),
value="birdy",
entity=slot_name,
slot_name=slot_name)
]
},
{
"text":
"light bird bird blue bird",
"tags": [
BEGINNING_PREFIX + slot_name, INSIDE_PREFIX + slot_name,
UNIT_PREFIX + slot_name, BEGINNING_PREFIX + slot_name,
INSIDE_PREFIX + slot_name
],
"expected_slots": [
unresolved_slot(match_range=(0, 10),
value="light bird",
entity=slot_name,
slot_name=slot_name),
unresolved_slot(match_range=(11, 15),
value="bird",
entity=slot_name,
slot_name=slot_name),
unresolved_slot(match_range=(16, 25),
value="blue bird",
entity=slot_name,
slot_name=slot_name)
]
},
{
"text":
"bird bird bird",
"tags": [
LAST_PREFIX + slot_name, BEGINNING_PREFIX + slot_name,
UNIT_PREFIX + slot_name
],
"expected_slots": [
unresolved_slot(match_range=(0, 4),
value="bird",
entity=slot_name,
slot_name=slot_name),
unresolved_slot(match_range=(5, 9),
value="bird",
entity=slot_name,
slot_name=slot_name),
unresolved_slot(match_range=(10, 14),
value="bird",
entity=slot_name,
slot_name=slot_name)
]
},
]
for data in tags:
# When
slots = tags_to_slots(data["text"],
tokenize(data["text"],
language), data["tags"],
TaggingScheme.BILOU, intent_slots_mapping)
# Then
self.assertEqual(slots, data["expected_slots"])
def test_entity_match_factory(self):
# Given
dataset_stream = io.StringIO("""
---
type: intent
name: my_intent
utterances:
- this is [entity1](my first entity)
- this is [entity2](second_entity)""")
dataset = Dataset.from_yaml_files("en", [dataset_stream]).json
config = {
"factory_name": "entity_match",
"args": {
"tagging_scheme_code": TaggingScheme.BILOU.value,
"use_stemming": True
},
"offsets": [0]
}
tokens = tokenize("my first entity and second_entity", LANGUAGE_EN)
cache = [{TOKEN_NAME: token} for token in tokens]
resources = {STEMS: dict()}
custom_entity_parser = CustomEntityParser.build(
dataset, CustomEntityParserUsage.WITH_STEMS, resources)
factory = CRFFeatureFactory.from_config(
config,
custom_entity_parser=custom_entity_parser,
resources=resources)
factory.fit(dataset, "my_intent")
# When
features = factory.build_features()
features = sorted(features, key=lambda f: f.base_name)
res0 = features[0].compute(0, cache)
res1 = features[0].compute(1, cache)
res2 = features[0].compute(2, cache)
res3 = features[0].compute(3, cache)
res4 = features[0].compute(4, cache)
res5 = features[1].compute(0, cache)
res6 = features[1].compute(1, cache)
res7 = features[1].compute(2, cache)
res8 = features[1].compute(3, cache)
res9 = features[1].compute(4, cache)
# Then
self.assertIsInstance(factory, CustomEntityMatchFactory)
self.assertEqual(len(features), 2)
self.assertEqual(features[0].base_name, "entity_match_entity1")
self.assertEqual(features[1].base_name, "entity_match_entity2")
self.assertEqual(res0, BEGINNING_PREFIX)
self.assertEqual(res1, INSIDE_PREFIX)
self.assertEqual(res2, LAST_PREFIX)
self.assertEqual(res3, None)
self.assertEqual(res4, None)
self.assertEqual(res5, None)
self.assertEqual(res6, None)
self.assertEqual(res7, None)
self.assertEqual(res8, None)
self.assertEqual(res9, UNIT_PREFIX)
