Extend spaCy prefixes
Improving tokenisation
import spacy
from spacy import util
batch_size = 100
nlp = spacy.blank("ga")
prefixes = nlp.Defaults.prefixes + [r"\([0-9]+\)", r"\$\d"]
prefix_regex = util.compile_prefix_regex(prefixes)
nlp.tokenizer.prefix_search = prefix_regex.search
nlp.add_pipe("sentencizer")
for doc in nlp.pipe(["Thosaigh sé. Bhí mé $2 $3.4 4% ag fanacht'm-ho at-ainm (9) théis IRP23 4ú 9pm ,000 1-0 3B ."], batch_size=batch_size):
for sent in doc.sents:
print(" ".join([t.text for t in sent]) + "\n")