安装语法依存包的时候出现问题

zixinzeng-jennifer · December 13, 2020, 12:43am

我在试图安装hanlp.pretrained.sdp.SEMEVAL15_PAS_BIAFFINE_EN（semantic dependency parsing工具）。
运行代码时，hanlp自动从网上下载了glove向量，但是似乎在unpack的时候遇到了错误。
在hanlp\components\parsers\conll.py 的489行form, cpos, head, deprel = sample[0]发生ValueError: not enough values to unpack (expected 4, got 3)的问题。
网上查找之后没有发生类似的帖子，求帮助orz

hankcs · December 13, 2020, 5:05pm

输入必须包括词性。

github.com

hankcs/HanLP/blob/fbbb38af3a70b7bfd05b384c0e5027a9cb9b36ab/tests/demo/en/demo_sdp.py

# -*- coding:utf-8 -*-
# Author: hankcs
# Date: 2020-01-03 15:26
import hanlp
from hanlp.components.parsers.conll import CoNLLSentence

# semeval15 offers three independent annotations over the Penn Treebank (PTB)
semantic_parser = hanlp.load(hanlp.pretrained.sdp.SEMEVAL15_PAS_BIAFFINE_EN)
# semantic_parser = hanlp.load(hanlp.pretrained.sdp.SEMEVAL15_DM_BIAFFINE_EN)
# semantic_parser = hanlp.load(hanlp.pretrained.sdp.SEMEVAL15_PSD_BIAFFINE_EN)
sent = [('Is', 'VBZ'),
        ('this', 'DT'),
        ('the', 'DT'),
        ('future', 'NN'),
        ('of', 'IN'),
        ('chamber', 'NN'),
        ('music', 'NN'),
        ('?', '.')]
tree = semantic_parser(sent)  # type:CoNLLSentence
print(tree)

zixinzeng-jennifer · December 14, 2020, 3:35pm

啊，非常感谢【我是没有传原文