def build_cnter(query: dict, *, verbose=False) -> Counter[str]:
[(tgt_key, tgt_val)] = query.items()
cnter = Counter()
for sentence in tqdm(mecab_into_sentences()):
cnter += Counter(d[tgt_key] for d in sentence)
if verbose:
with Renderer(f"「{tgt_val}」の出現頻度") as out:
out.header("上位 10 個")
pprint.pprint(cnter.most_common(10), stream=sys.stderr)
out.result("種類", len(cnter))
return cnter
https://nlp100.github.io/ja/ch04.html#33-aのb
[Usage]
python knock33.py
"""
import os
import sys
from typing import Dict, List
from knock30 import mecab_into_sentences
sys.path.append(os.path.join(os.path.dirname(__file__), "../../"))
from kiyuna.utils.message import Renderer # noqa: E402 isort:skip
Morpheme = Dict[str, str]
Sentence = List[Morpheme]
if __name__ == "__main__":
tgt = "AのB"
res = []
for sentence in mecab_into_sentences():
for a, no, b in zip(sentence, sentence[1:], sentence[2:]):
if (a["pos"], no["surface"], b["pos"]) == ("名詞", "の", "名詞"):
res.append("".join(map(lambda x: x["surface"], (a, no, b))))
with Renderer(tgt) as out:
out.result("数", len(res))
out.result("種類", len(set(res)))
out.result("上から 10 個", res[:10])