fix cov error

Author: statmlben

tests/test_mf.py

@@ -110,6 +110,169 @@ def test_mf_hinge_classification_fits(mf_data):
     assert accuracy > 0.5, f"Hinge-loss MF accuracy ({accuracy:.3f}) should be > 0.5"
 
 
+def test_mf_data_validation_errors():
+    """Test data validation raises appropriate errors."""
+    # Test X with wrong shape (not 2 columns)
+    with pytest.raises(ValueError, match="X must have shape"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"})
+        model.fit(np.array([[0, 0, 0]]), np.array([1.0]))
+
+    # Test X and y mismatch
+    with pytest.raises(ValueError, match="X and y must have the same number"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"})
+        model.fit(np.array([[0, 0]]), np.array([1.0, 2.0]))
+
+    # Test invalid user ID (negative)
+    with pytest.raises(ValueError, match="User IDs must be in"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"})
+        model.fit(np.array([[-1, 0]]), np.array([1.0]))
+
+    # Test invalid user ID (>= n_users)
+    with pytest.raises(ValueError, match="User IDs must be in"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"})
+        model.fit(np.array([[10, 0]]), np.array([1.0]))
+
+    # Test invalid item ID (negative)
+    with pytest.raises(ValueError, match="Item IDs must be in"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"})
+        model.fit(np.array([[0, -1]]), np.array([1.0]))
+
+    # Test invalid item ID (>= n_items)
+    with pytest.raises(ValueError, match="Item IDs must be in"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"})
+        model.fit(np.array([[0, 10]]), np.array([1.0]))
+
+
+def test_mf_cold_start_users_items():
+    """Test cold start handling: users/items with no interactions."""
+    # Create data where user 0 and item 0 have no interactions
+    # n_users=3, n_items=3, but only users 1,2 and items 1,2 interact
+    X = np.array([[1, 1], [1, 2], [2, 1], [2, 2]])
+    y = np.array([3.0, 4.0, 2.0, 5.0])
+
+    model = plqMF_Ridge(
+        n_users=3,
+        n_items=3,
+        loss={"name": "mae"},
+        rank=2,
+        C=0.1,
+        max_iter=1000,
+        tol=0.01,
+    )
+    model.fit(X, y)
+
+    # Cold start user (user 0) should have zero factors and bias
+    assert np.allclose(model.P[0, :], 0.0)
+    assert model.bu[0] == 0.0
+
+    # Cold start item (item 0) should have zero factors and bias
+    assert np.allclose(model.Q[0, :], 0.0)
+    assert model.bi[0] == 0.0
+
+
+def test_mf_biased_false():
+    """Test plqMF_Ridge with biased=False (no bias terms)."""
+    n_users, n_items = 20, 30
+    X = np.array([[0, 0], [0, 1], [1, 0], [1, 1], [2, 2], [3, 3]])
+    y = np.array([1.0, 2.0, 3.0, 4.0, 5.0, 6.0])
+
+    model = plqMF_Ridge(
+        n_users=n_users,
+        n_items=n_items,
+        loss={"name": "mae"},
+        biased=False,
+        rank=3,
+        C=0.1,
+        max_iter=1000,
+        tol=0.01,
+    )
+    model.fit(X, y)
+
+    # bu and bi should be None when biased=False
+    assert model.bu is None
+    assert model.bi is None
+
+    # decision_function should work without biases
+    scores = model.decision_function(X)
+    assert scores.shape == (len(X),)
+
+    # obj should work without biases
+    loss_term, obj_val = model.obj(X, y)
+    assert np.isfinite(loss_term)
+    assert np.isfinite(obj_val)
+
+
+def test_mf_verbose_output(capsys):
+    """Test verbose printing (lines 308, 464-466)."""
+    X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
+    y = np.array([1.0, 2.0, 3.0, 4.0])
+
+    # Test verbose=1 (CD iteration progress)
+    model = plqMF_Ridge(
+        n_users=2,
+        n_items=2,
+        loss={"name": "mae"},
+        rank=2,
+        C=0.1,
+        max_iter=500,
+        tol=0.01,
+        max_iter_CD=2,
+        verbose=1,
+    )
+    model.fit(X, y)
+    captured = capsys.readouterr()
+    assert "Iteration" in captured.out
+    assert "Average Loss" in captured.out
+
+
+def test_mf_convergence_warning():
+    """Test convergence warning when max_iter is too small."""
+    from sklearn.exceptions import ConvergenceWarning
+
+    X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
+    y = np.array([1.0, 2.0, 3.0, 4.0])
+
+    model = plqMF_Ridge(
+        n_users=2,
+        n_items=2,
+        loss={"name": "mae"},
+        rank=2,
+        C=0.1,
+        max_iter=1,  # Only 1 iteration to guarantee non-convergence
+        tol=1e-10,
+        max_iter_CD=1,
+    )
+    with pytest.warns(ConvergenceWarning, match="ReHLine failed to converge"):
+        model.fit(X, y)
+
+
+def test_mf_param_validation_errors():
+    """Test parameter validation raises appropriate errors."""
+    # Test invalid rho (must be between 0 and 1)
+    with pytest.raises(ValueError, match="rho must be between 0 and 1"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"}, rho=0.0)
+        model.fit(np.array([[0, 0]]), np.array([1.0]))
+
+    with pytest.raises(ValueError, match="rho must be between 0 and 1"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"}, rho=1.0)
+        model.fit(np.array([[0, 0]]), np.array([1.0]))
+
+    # Test invalid C (must be positive)
+    with pytest.raises(ValueError, match="C must be positive"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"}, C=0.0)
+        model.fit(np.array([[0, 0]]), np.array([1.0]))
+
+    # Test invalid tol_CD (must be positive)
+    with pytest.raises(ValueError, match="tol_CD must be positive"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"}, tol_CD=0.0)
+        model.fit(np.array([[0, 0]]), np.array([1.0]))
+
+    # Test invalid tol (must be positive)
+    with pytest.raises(ValueError, match="tol must be positive"):
+        model = plqMF_Ridge(n_users=10, n_items=10, loss={"name": "mae"}, tol=0.0)
+        model.fit(np.array([[0, 0]]), np.array([1.0]))
+
+
 def test_mf_nonneg_constraint(mf_data):
     """plqMF_Ridge with non-negative constraints should produce non-negative factors."""
     d = mf_data