What is the difference between pred_chunk_class and true_chunk_class？

[minskiter]

pchunk is the same as tchunk.

def evaluate_each_class(pred_chunks,true_chunks, class_type):
	# class_type:PER or LOC or ORG
	# pred_chunks,true_chunks are one-dim.
	pred_chunk_class = []
	for pchunk in pred_chunks:
		if pchunk[0] == class_type:
			pred_chunk_class.append(pchunk)

	true_chunk_class = []
	for tchunk in pred_chunks: # <- should be true_chunks
		if tchunk[0] == class_type:
			true_chunk_class.append(tchunk)
	pred_chunk_class = set(pred_chunk_class)
	true_chunk_class = set(true_chunk_class)



	correct_preds = len((pred_chunk_class & set(true_chunks) ))
	total_preds = len(pred_chunk_class)
	total_correct = len(true_chunk_class)
	# print("type: ", class_type)
	# print("correct_preds, total_preds, total_correct: ", correct_preds,total_preds,total_correct)

	p = correct_preds / total_preds if correct_preds > 0 else 0
	r = correct_preds / total_correct if correct_preds > 0 else 0
	f1 = 2 * p * r / (p + r) if correct_preds > 0 else 0

	return f1, p, r,correct_preds,total_preds,total_correct

[minskiter]

https://github.com/neulab/SpanNER/blob/main/combination/evaluate_metric.py
The calculation of recall is wrong.