Sparse matrix

[severinsimmler]

There is a pandas SparseDataFrame, which is interesting for our purposes especially with very large corpora. Someone should do some further research on the class. I've experimented with it a bit and found that some methods don't work like a normal DataFrame, and it all seems to me to be much slower.

See also:

cophi-toolbox/src/cophi/model.py

Lines 310 to 322 in 2c8a8f1

	class Corpus:
	"""Model class for a Corpus.
	Parameters:
	documents (iterable): One or more Document objects.
	sparse (str): If True, use the sparse DataFrame. NOT IMPLEMENTED.
	Attributes:
	dtm (pd.DataFrame): Document-term matrix with absolute word frequencies.
	"""
	def __init__(self, documents, sparse=False):
	if sparse:
	raise NotImplementedError("This feature is not yet implemented.")