如何自定义spaCy的标记器以排除正则表达式描述的拆分短语

import re
from spacy.tokenizer import Tokenizer
from spacy.util import compile_prefix_regex, compile_infix_regex, compile_suffix_regex

class CustomTokenizer(Tokenizer):
    def __init__(self, nlp):
        super().__init__(nlp)
        self.infixes = self.infixes + [r"(?<!\d)\.(?!\d)"]  # 添加一个新的中缀规则

def generate_infixes(exclude_phrases):
    infixes = []
    for phrase in exclude_phrases:
        infixes.append(r"(?<!" + re.escape(phrase) + r")\b")
    return infixes

nlp = spacy.load("en_core_web_sm")
exclude_phrases = ["正则表达式描述的拆分短语1", "正则表达式描述的拆分短语2"]  # 自定义的排除短语列表
infixes = generate_infixes(exclude_phrases)
custom_tokenizer = CustomTokenizer(nlp)
custom_tokenizer.infixes = custom_tokenizer.infixes + infixes
nlp.tokenizer = custom_tokenizer

# 使用自定义的标记器进行文本处理
doc = nlp("要处理的文本")