Example #1
0
def evaluate_chunker(treebank_root='corpora/treebank'):
	treebank = TreebankReader(treebank_root, join_clitics=True, join_verb_parts=True)
	chunker = Chunker()
	chunked_trees = list(treebank.chunked_trees())

	print(chunker.evaluate(chunked_trees))

	output = codecs.open('resources/chunker_errors.txt', 'w', 'utf8')
	for sentence, gold in zip(treebank.sents(), chunked_trees):
		chunked = chunker.parse(sentence)
		if chunked != gold:
			print(tree2brackets(chunked), file=output)
			print(tree2brackets(gold), file=output)
			print(file=output)
Example #2
0
File: data.py Project: zinaaa/hazm
def evaluate_chunker(treebank_root='corpora/treebank'):
	treebank = TreebankReader(treebank_root, join_clitics=True, join_verb_parts=True)
	chunker = Chunker()
	chunked_trees = list(treebank.chunked_trees())

	print(chunker.evaluate(chunked_trees))

	output = codecs.open('resources/chunker_errors.txt', 'w', 'utf8')
	for sentence, gold in zip(treebank.sents(), chunked_trees):
		chunked = chunker.parse(sentence)
		if chunked != gold:
			print(tree2brackets(chunked), file=output)
			print(tree2brackets(gold), file=output)
			print(file=output)