def test_get_information_content(self): ics = vsm.get_information_content(self.doc_term_matrix) self.assertEqual(len(ics), self.doc_term_matrix.shape[1]) self.assertAlmostEqual(ics.max(), 0.95443, places=4) self.assertAlmostEqual(ics.min(), 0.54356, places=4) self.assertAlmostEqual(ics[self.idx_lamb], 0.95443, places=4) self.assertAlmostEqual(ics[self.idx_child], 0.81127, places=4)
def test_get_information_content(self): ics = vsm.get_information_content(self.doc_term_matrix) self.assertEqual(len(ics), self.doc_term_matrix.shape[1]) self.assertAlmostEqual(ics.max(), 0.95443, places=4) self.assertAlmostEqual(ics.min(), 0.54356, places=4) self.assertAlmostEqual(ics[self.idx_lamb], 0.95443, places=4) self.assertAlmostEqual(ics[self.idx_child], 0.81127, places=4)
def test_get_information_content(vectorizer_and_dtm, lamb_and_child_idxs): _, doc_term_matrix = vectorizer_and_dtm idx_lamb, idx_child = lamb_and_child_idxs ics = vsm.get_information_content(doc_term_matrix) assert len(ics) == doc_term_matrix.shape[1] assert ics.max() == pytest.approx(1.0, rel=1e-3) assert ics.min() == pytest.approx(0.54356, rel=1e-3) assert ics[idx_lamb] == pytest.approx(0.95443, rel=1e-3) assert ics[idx_child] == pytest.approx(0.81127, rel=1e-3)
def test_get_information_content(vectorizer_and_dtm, lamb_and_children_idxs): _, doc_term_matrix = vectorizer_and_dtm idx_lamb, idx_children = lamb_and_children_idxs ics = vsm.get_information_content(doc_term_matrix) assert len(ics) == doc_term_matrix.shape[1] assert ((0.0 <= ics) & (ics <= 1.0)).all() assert ics.max() > ics.min() assert ics[idx_lamb] == pytest.approx(0.95443, rel=0.001) assert ics[idx_children] == pytest.approx(0.81127, rel=0.001)