def test_sparse_matrix(self):
rows = Rows(TestFeatureEncoder.rows)
fe = FeatureEncoder(rows)
inv_idx = fe.make_sparse_matrix()
expected = csr_matrix(((3, 2, 1, 3, 2, 1),
((0, 0, 0, 1, 1, 1), (0, 1, 2, 0, 1, 3))),
shape=(2, 4),
dtype=int32)
to_string = lambda mat: str(mat.toarray()).replace('\n', ',')
# scipy.sparse.matrix.nnz: number of non-zero values
self.assertEqual(abs(inv_idx - expected).nnz, 0,
"{} != {}".format(to_string(expected),
to_string(inv_idx)))
def test_sparse_matrix(self):
rows = Rows(TestFeatureEncoder.rows)
fe = FeatureEncoder(rows)
inv_idx = fe.make_sparse_matrix()
expected = csr_matrix(
((3, 2, 1, 3, 2, 1), ((0, 0, 0, 1, 1, 1), (0, 1, 2, 0, 1, 3))),
shape=(2, 4),
dtype=int32)
to_string = lambda mat: str(mat.toarray()).replace('\n', ',')
# scipy.sparse.matrix.nnz: number of non-zero values
self.assertEqual(
abs(inv_idx - expected).nnz, 0,
"{} != {}".format(to_string(expected), to_string(inv_idx)))
def test_make_vocabulary(self):
rows = Rows(TestFeatureEncoder.rows)
fe = FeatureEncoder(rows)
fe.make_sparse_matrix()
self.assertEqual({'a': 0, 'b': 1, 'c': 2, 'd': 3}, fe.vocabulary)
def test_make_vocabulary(self):
rows = Rows(TestFeatureEncoder.rows)
fe = FeatureEncoder(rows)
fe.make_sparse_matrix()
self.assertEqual({'a': 0, 'b': 1, 'c': 2, 'd': 3}, fe.vocabulary)