NUMBER is not supported entity type by spaCy

[HiromuHota]

Description of the bug

NumberMatcher matches spans if their ner_tag is either NUMBER or QUANTITY.

fonduer/src/fonduer/candidates/matchers.py

Lines 453 to 465 in 7f5c663

	class NumberMatcher(RegexMatchEach):
	"""
	Match Spans that are numbers, as identified by spaCy.
	A convenience class for setting up a RegexMatchEach to match spans
	for which each token was tagged as a number (NUMBER or QUANTITY).
	"""
	def __init__(self, *children, **kwargs): # type: ignore
	"""Initialize number matcher."""
	kwargs["attrib"] = "ner_tags"
	kwargs["rgx"] = "NUMBER|QUANTITY"
	super().__init__(*children, **kwargs)

However, NUMBER is not supported entity type by spaCy.
https://spacy.io/api/annotation

To Reproduce

N/A

Expected behavior

NumberMatcher should match if their ner_tag is either CARDINAL or QUANTITY.

This is an example result of spaCy as of v2.2,

>>> import spacy
>>> nlp = spacy.load("en")
>>> doc = nlp("He sold 100 million of iPhone.")
>>> for token in doc:
...     print(token.text, token.ent_type_)
... 
He 
sold 
100 CARDINAL
million CARDINAL
of 
iPhone ORG
. 

Error Logs/Screenshots

N/A

Environment (please complete the following information)

• Fonduer Version: 0.8.2

Additional context

The code was ported from Snorkel,

https://github.com/snorkel-team/snorkel-extraction/blob/d7faea18e4cbe0d26f3d0aef67839c6669556606/snorkel/matchers.py#L306-L316

where

Matches Spans that are numbers, as identified by CoreNLP.

CoreNLP uses NUMBER (https://stanfordnlp.github.io/CoreNLP/ner.html).