def load(cls, path):
import os
from pola.machine.topic_model.resource import FileResource
from pola.machine.topic_model.resource import PickleResource
ext = os.path.splitext(os.path.basename(path))[1].lower()
candidates = []
if ext == "txt":
r = FileResource(path)
candidates = r.load(deserializer=cls.__init__)
elif ext == "pickle":
r = PickleResource(path)
candidates = r.load()
return candidates
p = PickleResource(path)
doc = p.load()
if args.freq > 0:
doc.cut_frequent(args.freq)
doc.cut_pos({"pos": ["動詞", "副詞"], "class1": ["接尾", "副詞可能"], "class2": ["人名", "地域", "副詞可能"]})
if args.under > 0:
doc.cut_under(args.under)
if args.above > 0:
doc.cut_above(args.above)
if args.ignore:
ig_path = os.path.join(os.path.dirname(path), args.ignore)
ig = FileResource(ig_path)
words = ig.load()
for w in words:
doc.remove_vocab(w[0])
doc.show_vocab(show_pos=True)
if args.save:
fname = os.path.basename(path)
doc_fname = os.path.splitext(fname)[0] + "_edited.pickle"
doc_path = os.path.join(os.path.dirname(path), "./" + doc_fname)
pe = PickleResource(doc_path)
pe.save(doc)
