def test_breakups_extraction_2(self):
direction = Direction.BACKWARD
breakups = [
False, True, True, False, False, False, False, False, True, True,
False, False, False, False, True
]
self.assertEqual(
get_dt_breakups_feature(self.doc_with_1_sent, direction), breakups)
def test_breakups_extraction_3(self):
direction = Direction.FORWARD
sent_1_breakups = [
True, False, False, False, True, True, True, True, False, False,
True, True, True, True, False
]
sent_2_breakups = [
True, True, True, True, False, True, False, True, True, True, True,
False, False
]
breakups = sent_1_breakups + sent_2_breakups
self.assertEqual(
get_dt_breakups_feature(self.doc_with_2_sent, direction), breakups)
def _get_features(self, doc: Document) -> dict:
new_token_features = dict()
if "dt_head_distances" not in doc.token_features:
warn(
"SyntacticFeatureComputer was called on doc without dependency tree, some features won't be included"
)
else:
new_token_features["dt_depths"] = get_dt_depths_feature(doc)
for direction in [Direction.FORWARD, Direction.BACKWARD]:
new_token_features["dt_breakups_" +
direction.value] = get_dt_breakups_feature(
doc, direction=direction)
new_token_features[
"dt_deltas_" + direction.value] = get_dt_deltas_feature(
doc,
direction=direction,
precomputed_depths=new_token_features["dt_depths"])
new_token_features["borders"] = get_sentence_borders_feature(doc)
new_token_features.update(get_morph_features(doc, self.morph_features))
return new_token_features