def __init__(
self,
parse: bool = True,
rerank_model: PathLike = None,
asp_thresh: int = 3,
op_thresh: int = 2,
max_iter: int = 3,
):
self.acquire_lexicon = AcquireTerms(asp_thresh, op_thresh, max_iter)
if parse:
from nlp_architect.pipelines.spacy_bist import SpacyBISTParser
self.parser = SpacyBISTParser()
else:
self.parser = None
if not rerank_model:
print("using pre-trained reranking model")
rerank_model = _download_pretrained_rerank_model(
RERANK_MODEL_DEFAULT_PATH)
download_unzip(*EMBEDDING_URL,
EMBEDDING_PATH,
license_msg="Glove word embeddings.")
self.rerank = RerankTerms(vector_cache=True,
rerank_model=rerank_model,
emb_model_path=EMBEDDING_PATH)
def _download_pretrained_model():
"""Downloads the pre-trained BIST model if non-existent."""
if not path.isfile(SpacyBISTParser.dir / "bist.model"):
print("Downloading pre-trained BIST model..")
zip_path = SpacyBISTParser.dir / "bist-pretrained"
makedirs(SpacyBISTParser.dir, exist_ok=True)
download_unzip(
"https://d2zs9tzlek599f.cloudfront.net/models/dep_parse/",
"bist-pretrained.zip",
zip_path,
)
print("Done.")
def test_solution(generate_new=False):
lexicons_dir = Path(LIBRARY_ROOT) / "examples" / "absa"
expected_dir = Path(
LIBRARY_ROOT) / "tests" / "fixtures" / "data" / "absa_solution"
data_url = "https://d2zs9tzlek599f.cloudfront.net/tests/"
parsed_data = download_unzip(
data_url, "tripadvisor_test_parsed.zip",
SENTIMENT_OUT / "test" / "tripadvisor_test_parsed")
predicted_stats = SentimentSolution().run(
parsed_data=parsed_data,
aspect_lex=lexicons_dir / "aspects.csv",
opinion_lex=lexicons_dir / "opinions.csv",
)
predicted_stats.to_csv("predicted.csv", encoding="utf-8")
predicted_trimmed = pd.read_csv("predicted.csv",
encoding="utf-8").loc[:, "Aspect":"Score"]
predicted_trimmed.loc[:, "Score"] = np.around(
predicted_trimmed.loc[:, "Score"], 2)
os.remove("predicted.csv")
if generate_new:
with open("expected.csv", "w", encoding="utf-8", newline="") as f:
predicted_trimmed.to_csv(f)
assert False
else:
with open(expected_dir / "expected.csv",
encoding="utf-8") as expected_fp:
assert predicted_trimmed.to_csv() == expected_fp.read()
def _download_pretrained_model():
"""Downloads the pre-trained BIST model if non-existent."""
if not path.isfile(SpacyBISTParser.dir / "bist.model"):
print("Downloading pre-trained BIST model..")
zip_path = SpacyBISTParser.dir / "bist-pretrained"
makedirs(SpacyBISTParser.dir, exist_ok=True)
download_unzip(
"https://d2zs9tzlek599f.cloudfront.net/models/dep_parse/",
"bist-pretrained.zip",
zip_path,
)
# download_unlicensed_file(
# "https://d2zs9tzlek599f.cloudfront.net/models/dep_parse/",
# "bist-pretrained.zip",
# zip_path,
# )
# print("Unzipping...")
# uncompress_file(zip_path, outpath=str(SpacyBISTParser.dir))
# remove(zip_path)
print("Done.")
def test_solution():
lexicons_dir = Path(LIBRARY_ROOT) / 'examples' / 'absa' / 'inference'
expected_dir = Path(
LIBRARY_ROOT) / 'tests' / 'fixtures' / 'data' / 'absa_solution'
data_url = 'https://s3-us-west-2.amazonaws.com/nlp-architect-data/tests/'
parsed_data = download_unzip(
data_url, 'tripadvisor_test_parsed.zip',
SENTIMENT_OUT / 'test' / 'tripadvisor_test_parsed')
predicted_stats = SentimentSolution().run(
parsed_data=parsed_data,
aspect_lex=lexicons_dir / 'aspects.csv',
opinion_lex=lexicons_dir / 'opinions.csv',
ui=False)
predicted_stats.to_csv('predicted.csv', encoding='utf-8')
predicted_trimmed = pd.read_csv('predicted.csv',
encoding='utf-8').loc[:, 'Aspect':'Score']
predicted_trimmed.loc[:, 'Score'] = np.around(
predicted_trimmed.loc[:, 'Score'], 2)
os.remove('predicted.csv')
with open(expected_dir / 'expected.csv', encoding='utf-8') as expected_fp:
assert predicted_trimmed.to_csv() == expected_fp.read()
