wmkouw /
libTLDA
| @@ 73-82 (lines=10) @@ | ||
| 70 | X = rnd.randn(10, 2) |
|
| 71 | y = np.hstack((-np.ones((5,)), np.ones((5,)))) |
|
| 72 | Z = rnd.randn(10, 2) + 1 |
|
| 73 | clf = ImportanceWeightedClassifier() |
|
| 74 | clf.fit(X, y, Z) |
|
| 75 | u_pred = clf.predict(Z) |
|
| 76 | labels = np.unique(y) |
|
| 77 | assert len(np.setdiff1d(np.unique(u_pred), labels)) == 0 |
|
| 78 | ||
| @@ 28-37 (lines=10) @@ | ||
| 25 | """Test for fitting the model.""" |
|
| 26 | X = rnd.randn(10, 2) |
|
| 27 | y = np.hstack((-np.ones((5,)), np.ones((5,)))) |
|
| 28 | Z = rnd.randn(10, 2) + 1 |
|
| 29 | clf = SubspaceAlignedClassifier() |
|
| 30 | clf.fit(X, y, Z) |
|
| 31 | assert clf.is_trained |
|
| 32 | ||
| 33 | ||
| 34 | def test_predict(): |
|
| 35 | """Test for making predictions.""" |
|
| 36 | X = rnd.randn(10, 2) |
|
| 37 | y = np.hstack((-np.ones((5,)), np.ones((5,)))) |
|
| 38 | Z = rnd.randn(10, 2) + 1 |
|
| 39 | clf = SubspaceAlignedClassifier() |
|
| 40 | clf.fit(X, y, Z) |
|
