test_devnet.py
1 # -*- coding: utf-8 -*- 2 from __future__ import division 3 from __future__ import print_function 4 5 import os 6 import sys 7 import unittest 8 9 import numpy as np 10 import torch 11 from numpy.testing import assert_almost_equal 12 from numpy.testing import assert_equal 13 from numpy.testing import assert_raises 14 from sklearn.metrics import roc_auc_score 15 16 # temporary solution for relative imports in case pyod is not installed 17 # if pyod is installed, no need to use the following line 18 19 sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..'))) 20 21 from pyod.models.devnet import DevNet 22 from pyod.utils.data import generate_data 23 24 25 class TestDevNet(unittest.TestCase): 26 def setUp(self): 27 self.n_train = 3000 28 self.n_test = 1500 29 self.n_features = 2000 30 self.contamination = 0.1 31 self.roc_floor = 0.8 32 self.X_train, self.X_test, self.y_train, self.y_test = generate_data( 33 n_train=self.n_train, n_test=self.n_test, 34 n_features=self.n_features, contamination=self.contamination, 35 random_state=42) 36 37 self.clf = DevNet(epochs=3, contamination=self.contamination) 38 self.clf.fit(self.X_train, self.y_train) 39 40 def test_parameters(self): 41 assert (hasattr(self.clf, 'decision_scores_') and 42 self.clf.decision_scores_ is not None) 43 assert (hasattr(self.clf, 'labels_') and 44 self.clf.labels_ is not None) 45 assert (hasattr(self.clf, 'threshold_') and 46 self.clf.threshold_ is not None) 47 assert (hasattr(self.clf, '_mu') and 48 self.clf._mu is not None) 49 assert (hasattr(self.clf, '_sigma') and 50 self.clf._sigma is not None) 51 assert (hasattr(self.clf, 'model') and 52 self.clf.model is not None) 53 54 def test_train_scores(self): 55 assert_equal(len(self.clf.decision_scores_), self.X_train.shape[0]) 56 57 def test_prediction_scores(self): 58 pred_scores = self.clf.decision_function(self.X_test) 59 60 # check score shapes 61 assert_equal(pred_scores.shape[0], self.X_test.shape[0]) 62 63 # check performance 64 assert (roc_auc_score(self.y_test, pred_scores) >= self.roc_floor) 65 66 def test_prediction_labels(self): 67 pred_labels = self.clf.predict(self.X_test) 68 assert_equal(pred_labels.shape, self.y_test.shape) 69 70 def test_prediction_proba(self): 71 pred_proba = self.clf.predict_proba(self.X_test) 72 assert (pred_proba.min() >= 0) 73 assert (pred_proba.max() <= 1) 74 75 def test_prediction_proba_linear(self): 76 pred_proba = self.clf.predict_proba(self.X_test, method='linear') 77 assert (pred_proba.min() >= 0) 78 assert (pred_proba.max() <= 1) 79 80 def test_prediction_proba_unify(self): 81 pred_proba = self.clf.predict_proba(self.X_test, method='unify') 82 assert (pred_proba.min() >= 0) 83 assert (pred_proba.max() <= 1) 84 85 def test_prediction_proba_parameter(self): 86 with assert_raises(ValueError): 87 self.clf.predict_proba(self.X_test, method='something') 88 89 def test_prediction_labels_confidence(self): 90 pred_labels, confidence = self.clf.predict(self.X_test, 91 return_confidence=True) 92 assert_equal(pred_labels.shape, self.y_test.shape) 93 assert_equal(confidence.shape, self.y_test.shape) 94 assert (confidence.min() >= 0) 95 assert (confidence.max() <= 1) 96 97 def test_prediction_proba_linear_confidence(self): 98 pred_proba, confidence = self.clf.predict_proba(self.X_test, 99 method='linear', 100 return_confidence=True) 101 assert (pred_proba.min() >= 0) 102 assert (pred_proba.max() <= 1) 103 104 assert_equal(confidence.shape, self.y_test.shape) 105 assert (confidence.min() >= 0) 106 assert (confidence.max() <= 1) 107 108 def test_prediction_with_rejection(self): 109 pred_labels = self.clf.predict_with_rejection(self.X_test, 110 return_stats=False) 111 assert_equal(pred_labels.shape, self.y_test.shape) 112 113 def test_prediction_with_rejection_stats(self): 114 _, [expected_rejrate, ub_rejrate, 115 ub_cost] = self.clf.predict_with_rejection(self.X_test, 116 return_stats=True) 117 assert (expected_rejrate >= 0) 118 assert (expected_rejrate <= 1) 119 assert (ub_rejrate >= 0) 120 assert (ub_rejrate <= 1) 121 assert (ub_cost >= 0) 122 123 def test_fit_predict(self): 124 pred_labels = self.clf.fit_predict(self.X_train, self.y_train) 125 assert_equal(pred_labels.shape, self.y_train.shape) 126 127 def test_fit_predict_score(self): 128 self.clf.fit_predict_score(self.X_test, self.y_test) 129 self.clf.fit_predict_score(self.X_test, self.y_test, 130 scoring='roc_auc_score') 131 self.clf.fit_predict_score(self.X_test, self.y_test, 132 scoring='prc_n_score') 133 with assert_raises(NotImplementedError): 134 self.clf.fit_predict_score(self.X_test, self.y_test, 135 scoring='something') 136 137 def test_model_clone(self): 138 pass 139 # clone_clf = clone(self.clf) 140 141 def tearDown(self): 142 pass 143 144 145 if __name__ == '__main__': 146 unittest.main()