KaneAI - World’s First E2E Software Testing Agent.Book a Demo 
How to use test_pattern method in lisa
Best Python code snippet using lisa_python
test_feedforward.py
Source:test_feedforward.py 
1import numpy as np2from BCPNN.feedforward import BCPNN3from BCPNN.encoder import BinvecOneHotEncoder as encoder4class TestUnitTests:5 def setup(self):6 self.clf = BCPNN()7 self.test_pattern = np.array([8 [1, 0, 1, 1, 0],9 [1, 1, 0, 1, 0]10 ])11 self.transformed_pattern = encoder.transform(self.test_pattern)12 self.y = np.array([0,1])13 self.clf.fit(self.test_pattern, self.y)14 def test_encoder_encode(self):15 transformed = np.array([1, 0, 0, 1, 1, 0, 1, 0, 0, 1])16 assert (self.transformed_pattern[0] == transformed).all()17 def test_encoder_decode(self):18 inverse = encoder.inverse_transform(self.transformed_pattern)19 assert (inverse == self.test_pattern).all()20 def test_get_prob(self):21 f = self.clf._get_prob22 assert f(0) == 123 assert f(1) == 0.524 assert f(2) == 0.525 assert f(3) == 126 assert f(4) == 027 def test_get_joint_prob(self):28 f = self.clf._get_joint_prob29 assert f(0, 0) == 130 assert f(0, 1) == 0.531 assert f(1, 4) == 032 def test_get_prob_for_y(self):33 f = self.clf._get_prob34 assert f(self.clf.y_pad + 0) == 0.535 assert f(self.clf.y_pad + 1) == 0.536 def test_get_joint_prob_for_y(self):37 f = self.clf._get_joint_prob38 assert f(0, self.clf.y_pad + 0) == 0.539 assert f(0, self.clf.y_pad + 1) == 0.540 assert f(1, self.clf.y_pad + 1) == 0.541 assert f(1, self.clf.y_pad + 0) == 042 assert f(4, self.clf.y_pad + 1) == 043 def test_unique_label(self):44 y = [3,4,1,5,6,7,2,2,3]45 assert (BCPNN._unique_labels(y) == [1,2,3,4,5,6,7]).all()46 def test_class_idx_to_prob(self):47 y = np.arange(3)48 prediction = np.eye(3)49 assert (BCPNN._class_idx_to_prob(y) == prediction).all()50 y = np.array([1, 2, 0])51 prediction = np.array([[0, 1, 0], [0, 0, 1], [1, 0, 0]])52 assert (BCPNN._class_idx_to_prob(y) == prediction).all()53 y = np.array([1, 1, 0])54 prediction = np.array([[0, 1], [0,1], [1, 0]])55 assert (BCPNN._class_idx_to_prob(y) == prediction).all()56 def test_transfer_fn(self):57 g = self.clf._transfer_fn58 f = lambda s: g(np.array(s))59 support = [0, 0]60 assert (f(support) == [1, 1]).all()61 support = np.log([1, 2, 4, 1])62 assert (f(support) == [1, 1, 1, 1]).all()63 # test 2d arrays64 support = [[0, 0], [0, 0]]65 assert (f(support) == [[1,1], [1,1]]).all()66 support = np.log([[2, 6], [4, 1]]).tolist()67 assert (f(support) == [[1,1], [1,1]]).all()68class TestProba:69 test_pattern = np.array([70 [1, 0, 1, 0, 0, 1],71 [1, 0, 0, 1, 0, 1]72 ])73 targets = np.array([0, 1])74 def clf_factory(self, test_pattern=test_pattern, targets=targets):75 clf = BCPNN()76 clf.fit(test_pattern, targets)77 return clf78 def predict_runner(self, train_pattern, targets, predictions,79 test_pattern=None, mode='proba', atol=0.2):80 clf = self.clf_factory(train_pattern, targets)81 if mode == 'proba':82 f = clf.predict_proba83 elif mode == 'log':84 f = clf.predict_log_proba85 elif mode == 'predict':86 f = clf.predict87 if test_pattern is None:88 test_pattern = train_pattern89 output = f(test_pattern)90 assert output.shape == predictions.shape91 # NOTE: below form easier to debug92 # assert (output == predictions).all()93 assert np.allclose(output, predictions, atol=atol)94 def test_basic1(self):95 test_pattern = np.array([96 [1, 0, 1, 0],97 [0, 1, 0, 1]98 ])99 targets = np.array([0, 1])100 # predict_log_proba101 predictions = np.array([[ 0.69314718, -2.07944154],102 [-2.07944154, 0.69314718]])103 self.predict_runner(test_pattern, targets, predictions,104 mode='log', atol=0)105 # predict_proba106 predictions = np.array([[1, 0], [0, 1]])107 self.predict_runner(test_pattern, targets, predictions, mode='proba')108 # predict109 predictions = np.array([0, 1])110 self.predict_runner(test_pattern, targets, predictions, mode='predict')111 def test_basic2(self):112 train_pattern = np.array([113 [1, 0, 1, 0],114 [0, 1, 0, 1]115 ])116 targets = np.array([0, 1])117 test_pattern = np.append(train_pattern, [[1, 1, 1, 1]], axis=0)118 # predict_proba119 predictions = np.array([[1, 0], [0, 1], [0.5, 0.5]])120 self.predict_runner(train_pattern, targets, predictions,121 test_pattern=test_pattern, mode='proba')122 # predict123 predictions = np.array([0, 1, 0])124 self.predict_runner(train_pattern, targets, predictions,125 test_pattern=test_pattern, mode='predict')126 def test_basic3(self):127 train_pattern = np.array([128 [1, 0, 0],129 [0, 1, 0],130 ])131 targets = np.array([0, 1])132 test_pattern = np.append(train_pattern, [[1, 1, 1], [1, 1, 0]], axis=0)133 # predict_proba134 predictions = np.array([[1, 0.25], [0.25, 1], [0.5, 0.5], [0.5, 0.5]])135 self.predict_runner(train_pattern, targets, predictions,136 test_pattern=test_pattern, mode='proba')137 # predict138 predictions = np.array([0, 1, 0, 0])139 self.predict_runner(train_pattern, targets, predictions,140 test_pattern=test_pattern, mode='predict')141 def test_basic3_2(self):142 train_pattern = np.array([143 [1, 0, 1],144 [0, 1, 0],145 # we add one more sample of cls 1, as for both classes to146 # have equal amount of active features147 [0, 1, 0]148 ])149 targets = np.array([0, 1, 1])150 test_pattern = np.append(train_pattern, [[1, 1, 1], [1, 1, 0]], axis=0)151 # predict_proba152 predictions = np.array([[1, 0], [0, 1], [0, 1], [0.8, 0.2], [0.5, 0.5]])153 self.predict_runner(train_pattern, targets, predictions,154 test_pattern=test_pattern, mode='proba')155 # predict156 predictions = np.array([0, 1, 1, 0, 0])157 self.predict_runner(train_pattern, targets, predictions,158 test_pattern=test_pattern, mode='predict')159 def test_basic4(self):160 train_pattern = np.array([161 [1, 0, 0],162 [0, 1, 0]163 ])164 targets = np.array([0, 1])165 # TEST CASE166 test_pattern = np.append(train_pattern, [[0, 0, 1]], axis=0)167 # predict_proba168 predictions = np.array([[1, 0], [0, 1], [0.5, 0.5]])169 self.predict_runner(train_pattern, targets, predictions,170 test_pattern=test_pattern, mode='proba', atol=0.25)171 # predict172 predictions = np.array([0, 1, 0])173 self.predict_runner(train_pattern, targets, predictions,174 test_pattern=test_pattern, mode='predict')175 # TEST CASE176 test_pattern = np.append(train_pattern, [[0, 0, 0]], axis=0)177 # predict_proba178 predictions = np.array([[1, 0], [0, 1], [0.5, 0.5]])179 self.predict_runner(train_pattern, targets, predictions,180 test_pattern=test_pattern, mode='proba', atol=0.25)181 # predict182 predictions = np.array([0, 1, 0])183 self.predict_runner(train_pattern, targets, predictions,184 test_pattern=test_pattern, mode='predict')185 def test_basic5(self):186 train_pattern = np.array([187 [1, 0, 0],188 [0, 1, 0],189 [0, 0, 1]190 ])191 targets = np.array([0, 1, 2])192 # TEST CASE193 test_pattern = np.append(train_pattern, [[0, 0, 0]], axis=0)194 # predict_proba195 predictions = np.array([[1, 0, 0], [0, 1, 0], [0, 0, 1], [0.5, 0.5, 0.5]])196 self.predict_runner(train_pattern, targets, predictions,197 test_pattern=test_pattern, mode='proba', atol=0.2)198 # predict199 predictions = np.array([0, 1, 2, 0])200 self.predict_runner(train_pattern, targets, predictions,201 test_pattern=test_pattern, mode='predict')202 # TEST CASE203 test_pattern = np.append(train_pattern, [[1, 1, 1]], axis=0)204 # predict_proba205 predictions = np.array([[1, 0, 0], [0, 1, 0], [0, 0, 1], [0.3, 0.3, 0.3]])206 self.predict_runner(train_pattern, targets, predictions,207 test_pattern=test_pattern, mode='proba', atol=0.2)208 # TEST CASE209 test_pattern = np.append(train_pattern, [[1, 0, 1]], axis=0)210 # predict_proba211 predictions = np.array([[1, 0, 0], [0, 1, 0], [0, 0, 1], [0.3, 0, 0.3]])212 self.predict_runner(train_pattern, targets, predictions,213 test_pattern=test_pattern, mode='proba', atol=0.2)214 def test_different_sizes1(self):215 # Test different size of n_features and n_classes216 # n_features = 5, n_classes = 2217 test_pattern = np.array([218 [1, 0, 1, 0, 1],219 [1, 0, 1, 0, 1],220 [0, 1, 0, 1, 0]221 ])222 targets = np.array([0, 0, 1])223 # predict_proba224 predictions = np.array([[1, 0], [1, 0], [0, 1]])225 self.predict_runner(test_pattern, targets, predictions, mode='proba')226 # predict227 predictions = np.array([0, 0, 1])228 self.predict_runner(test_pattern, targets, predictions, mode='predict')229 def test_different_sizes2(self):230 # Test different size of n_features and n_classes231 # n_features = 4, n_classes = 3232 test_pattern = np.array([233 [1, 0, 0],234 [0, 1, 0],235 [0, 0, 1]236 ])237 targets = np.array([0, 1, 2])238 # predict_proba239 predictions = np.array([[1, 0, 0], [0, 1, 0], [0, 0, 1]])240 self.predict_runner(test_pattern, targets, predictions, mode='proba')241 # predict242 predictions = np.array([0, 1, 2])243 self.predict_runner(test_pattern, targets, predictions, mode='predict')244 def test_different_sizes3(self):245 # Test different size of n_features and n_classes246 # n_features = 3, n_classes = 3247 test_pattern = np.array([248 [1, 0, 1],249 [1, 0, 1],250 [0, 1, 0]251 ])252 targets = np.array([0, 2, 1])253 # predict_proba254 predictions = np.array([[0.75 , 0.03703704, 0.75 ],255 [0.75 , 0.03703704, 0.75 ],256 [0.11111111, 1. , 0.11111111]])257 self.predict_runner(test_pattern, targets, predictions, mode='proba')258 # predict259 predictions = np.array([0, 0, 1])260 self.predict_runner(test_pattern, targets, predictions, mode='predict')261 def test_different_sizes4(self):262 # Test different size of n_features and n_classes263 # n_features = 4, n_classes = 3264 test_pattern = np.array([265 [1, 0, 1, 0, 0, 0, 0],266 [0, 0, 0, 0, 1, 1, 1],267 [0, 1, 0, 1, 0, 0, 0]268 ])269 targets = np.array([0, 1, 2])270 # predict_proba271 predictions = np.array([[1, 0, 0], [0, 1, 0], [0, 0, 1]])272 self.predict_runner(test_pattern, targets, predictions, mode='proba')273 # predict274 predictions = np.array([0, 1, 2])275 self.predict_runner(test_pattern, targets, predictions, mode='predict')276 # Encoder tests277 def test_encoder_effect_1(self):278 # extension of basic_3279 train_pattern = np.array([280 [1, 0, 1],281 [0, 1, 0]282 ])283 targets = np.array([0, 1])284 test_pattern = np.append(train_pattern, [[1, 1, 1], [1, 1, 0]], axis=0)285 train_pattern = encoder.transform(train_pattern)286 test_pattern = encoder.transform(test_pattern)287 # predict_proba288 predictions = np.array([289 [1. , 0.0625],290 [0.0625, 1. ],291 [1. , 0.25 ],292 [0.25 , 1. ]])293 self.predict_runner(train_pattern, targets, predictions,294 test_pattern=test_pattern, mode='proba')295 # predict296 predictions = np.array([0, 1, 0, 1])297 self.predict_runner(train_pattern, targets, predictions,298 test_pattern=test_pattern, mode='predict')299 def test_encoder_effect_2(self):300 # extension of test_different_sizes3301 test_pattern = np.array([302 [1, 0, 1],303 [1, 0, 1],304 [0, 1, 0]305 ])306 targets = np.array([0, 2, 1])307 # train_pattern = encoder.transform(train_pattern)308 test_pattern = encoder.transform(test_pattern)309 # predict_proba310 predictions = np.array([311 [1. , 0.01234568, 1. ],312 [1. , 0.01234568, 1. ],313 [0.01234568, 1. , 0.01234568]])314 self.predict_runner(test_pattern, targets, predictions, mode='proba')315 # predict316 predictions = np.array([0, 0, 1])317 self.predict_runner(test_pattern, targets, predictions, mode='predict')318 def test_encoder_effect_3(self):319 train_pattern = np.array([320 [1, 0, 0],321 [0, 1, 0],322 [0, 0, 1]323 ])324 targets = np.array([0, 1, 2])325 test_pattern = np.append(train_pattern, [[1, 1, 1], [1, 1, 0]], axis=0)326 train_pattern = encoder.transform(train_pattern)327 test_pattern = encoder.transform(test_pattern)328 # predict_proba329 predictions = np.array([[1. , 0.05555556, 0.05555556],330 [0.05555556, 1. , 0.05555556],331 [0.05555556, 0.05555556, 1. ],332 [0.11111111, 0.11111111, 0.11111111],333 [0.5 , 0.5 , 0.01234568]])334 self.predict_runner(train_pattern, targets, predictions,335 test_pattern=test_pattern, mode='proba')336 # predict337 predictions = np.array([0, 1, 2, 0, 0])338 self.predict_runner(train_pattern, targets, predictions,...
test_modular_ff.py
Source:test_modular_ff.py
1import pytest2import numpy as np3from BCPNN.feedforward_modular import mBCPNN as BCPNN4from BCPNN.encoder import ComplementEncoder5## TEST UTILS6def clf_factory(test_pattern, targets, module_sizes, normalize=True):7 clf = BCPNN(normalize=normalize)8 clf.fit(test_pattern, targets, module_sizes)9 return clf10def predict_runner(train_pattern, targets, predictions, module_sizes,11 test_pattern=None, mode='proba', atol=0.001, normalize=True):12 clf = clf_factory(train_pattern, targets, module_sizes, normalize)13 if mode == 'proba':14 f = clf.predict_proba15 elif mode == 'log':16 f = clf.predict_log_proba17 elif mode == 'predict':18 f = clf.predict19 if test_pattern is None:20 test_pattern = train_pattern21 output = f(test_pattern)22 assert output.shape == predictions.shape23 # NOTE: below form easier to debug24 # assert (output == predictions).all()25 assert np.allclose(output, predictions, atol=atol)26 assert (clf.weights != 0).all()27class TestUnitTests:28 def testComplementEncoder(self):29 X = np.array([[0.5, 0.2], [1, 0]])30 Xt = ComplementEncoder().fit_transform(X)31 assert np.array_equal(Xt, np.array([[0.5, 0.5, 0.2, 0.8], [1, 0, 0, 1]]))32 assert np.array_equal(ComplementEncoder.inverse_transform(Xt), X)33 @pytest.fixture(scope="function")34 def clf(self):35 return BCPNN()36 def testIndexTransform(self, clf):37 clf.module_sizes = np.array([1])38 modular = (0, 0)39 flat = 040 assert clf._modular_idx_to_flat(*modular) == flat41 assert clf._flat_to_modular_idx(flat) == modular42 clf.module_sizes = np.array([1, 1])43 modular = (1, 0)44 flat = 145 assert clf._modular_idx_to_flat(*modular) == flat46 assert clf._flat_to_modular_idx(flat) == modular47 clf.module_sizes = np.array([2, 2])48 modular = (0, 1)49 flat = 150 assert clf._modular_idx_to_flat(*modular) == flat51 assert clf._flat_to_modular_idx(flat) == modular52 modular = (1, 0)53 flat = 254 assert clf._modular_idx_to_flat(*modular) == flat55 assert clf._flat_to_modular_idx(flat) == modular56 clf.module_sizes = np.array([1, 3, 2])57 modular = (2, 1)58 flat = 559 assert clf._modular_idx_to_flat(*modular) == flat60 assert clf._flat_to_modular_idx(flat) == modular61 def testEmptyModuleSize(self, clf):62 X = np.array([[0, 1]])63 y = np.array([0])64 clf.fit(X, y)65 assert np.array_equal(clf.X_, X)66 assert np.array_equal(clf.module_sizes, [2, 1])67 class TestNormalization:68 def testCheckNormalization1(self, clf):69 X = np.array([[0, 1]])70 module_sizes = np.array([2, 2])71 clf.fit(X, X, module_sizes)72 X = np.array([[1, 1]])73 with pytest.raises(BCPNN.NormalizationError):74 clf.fit(X, X, module_sizes)75 def testCheckNormalization2(self, clf):76 X = np.array([[0, 1], [1/2, 1/2]])77 module_sizes = np.array([2, 2])78 clf.fit(X, X, module_sizes)79 clf._assert_module_normalization(X)80 X = np.array([[1, 1], [1/2, 1/2]])81 module_sizes = np.array([2, 2])82 with pytest.raises(BCPNN.NormalizationError):83 clf.fit(X, X, module_sizes)84 def testCheckNormalization3(self, clf):85 X = np.array([[0, 1/2, 1/2, 1/2, 1/2]])86 module_sizes = np.array([3, 2, 2])87 clf.fit(X, X[:, :2], module_sizes)88 X = np.array([[1, 1, 1, 1, 1]])89 module_sizes = np.array([3, 2, 2])90 with pytest.raises(BCPNN.NormalizationError):91 clf.fit(X, X[:, :2], module_sizes)92 def testCheckNormalization4(self, clf):93 # No module sizes given94 X = np.array([[0, 1]])95 clf.fit(X, X)96 X = np.array([[1, 1]])97 with pytest.raises(BCPNN.NormalizationError):98 clf.fit(X, X)99class TestModule:100 class TestBinary:101 def testModuleSize2_2_log(self):102 train_pattern = np.array([[0, 1], [1, 0]])103 targets = np.array([[0, 1], [1, 0]])104 predictions = np.array([[np.log(1/4), 0], [0, np.log(1/4)]])105 module_sizes = np.array([2, 2])106 predict_runner(train_pattern, targets, predictions, module_sizes, mode='log')107 def testModuleSize2_2_predict(self):108 train_pattern = np.array([[0, 1], [1, 0]])109 targets = np.array([[0, 1], [1, 0]])110 predictions = np.array([1, 0])111 module_sizes = np.array([2, 2])112 predict_runner(train_pattern, targets, predictions, module_sizes, mode='predict')113 def testModuleSize2_2_proba_no_norm(self):114 train_pattern = np.array([[0, 1], [1, 0]])115 targets = np.array([[0, 1], [1, 0]])116 predictions = np.array([[0.25, 1], [1, 0.25]])117 module_sizes = np.array([2, 2])118 predict_runner(train_pattern, targets, predictions, module_sizes, mode='proba', normalize=False)119 def testModuleSize2_2_proba(self):120 train_pattern = np.array([[0, 1], [1, 0]])121 targets = np.array([[0, 1], [1, 0]])122 predictions = np.array([[0.2, 0.8], [0.8, 0.2]])123 module_sizes = np.array([2, 2])124 predict_runner(train_pattern, targets, predictions, module_sizes, mode='proba')125 def testModuleSize3_2_proba(self):126 train_pattern = np.array([[0, 1, 0], [1, 0, 0]])127 targets = np.array([[0, 1], [1, 0]])128 predictions = np.array([[0.2, 0.8], [0.8, 0.2]])129 module_sizes = np.array([3, 2])130 predict_runner(train_pattern, targets, predictions, module_sizes, mode='proba')131 def testModuleSize3_3_log(self):132 train_pattern = np.array([[0, 1, 0], [1, 0, 0]])133 targets = np.array([[0, 1, 0], [1, 0, 0]])134 predictions = np.array([[np.log(1/4), 0, np.log(1/4)], [0, np.log(1/4), np.log(1/4)]])135 module_sizes = np.array([3, 3])136 predict_runner(train_pattern, targets, predictions, module_sizes, mode='log')137 def testModuleSize3_3_proba(self):138 train_pattern = np.array([[0, 1, 0], [1, 0, 0]])139 targets = np.array([[0, 1, 0], [1, 0, 0]])140 predictions = np.array([[1/6, 2/3, 1/6], [2/3, 1/6, 1/6]])141 module_sizes = np.array([3, 3])142 predict_runner(train_pattern, targets, predictions, module_sizes, mode='proba')143 def testModuleSize2_2_2_log(self):144 train_pattern = np.array([[0, 1, 0, 1], [1, 0, 1, 0]])145 targets = np.array([[0, 1], [1, 0]])146 predictions = np.array([[np.log(1/8), np.log(2)], [np.log(2), np.log(1/8)]])147 module_sizes = np.array([2, 2, 2])148 predict_runner(train_pattern, targets, predictions, module_sizes, mode='log')149 def testModuleSize2_2_2_proba(self):150 train_pattern = np.array([[0, 1, 0, 1], [1, 0, 1, 0]])151 targets = np.array([[0, 1], [1, 0]])152 predictions = np.array([[1/17, 16/17], [16/17, 1/17]])153 module_sizes = np.array([2, 2, 2])154 predict_runner(train_pattern, targets, predictions, module_sizes, mode='proba')155 def testModuleSize2_3_2_log(self):156 train_pattern = np.array([[0, 1, 1, 0, 0], [1, 0, 0, 1, 0]])157 targets = np.array([[0, 1], [1, 0]])158 predictions = np.array([[np.log(1/8), np.log(2)], [np.log(2), np.log(1/8)]])159 module_sizes = np.array([2, 3, 2])160 predict_runner(train_pattern, targets, predictions, module_sizes, mode='log')161 def testModuleNormalizationAssertion1_train(self):162 train_pattern = np.array([[1, 1]])163 targets = np.array([[1, 0]])164 test_pattern = np.array([[0, 0]])165 predictions = np.array([[1, 0]])166 module_sizes = np.array([2, 2])167 with pytest.raises(BCPNN.NormalizationError):168 predict_runner(train_pattern, targets, predictions, module_sizes, test_pattern=test_pattern,169 mode='proba')170 def testModuleNormalizationAssertion1_test(self):171 train_pattern = np.array([[1, 0]])172 targets = np.array([[1, 0]])173 test_pattern = np.array([[1, 1]])174 predictions = np.array([[1, 0]])175 module_sizes = np.array([2, 2])176 with pytest.raises(BCPNN.NormalizationError):177 predict_runner(train_pattern, targets, predictions, module_sizes, test_pattern=test_pattern,178 mode='proba')179 class TestFractional:180 def testModuleSize2_2_log(self):181 train_pattern = np.array([[1/3, 2/3], [2/3, 1/3]])182 targets = np.array([[0, 1], [1, 0]])183 predictions = np.array([[np.log(4/9), np.log(5/9)], [np.log(5/9), np.log(4/9)]])184 module_sizes = np.array([2, 2])185 predict_runner(train_pattern, targets, predictions, module_sizes, mode='log')186 def testModuleSize2_3_2_log(self):187 train_pattern = np.array([[1/3, 2/3, 1/4, 1/4, 2/4], [1, 0, 0, 1, 0]])188 targets = np.array([[0, 1], [1, 0]])189 predictions = np.array([[np.log(155/480), np.log(288/240)], [np.log(6/5), np.log(1/10)]])190 module_sizes = np.array([2, 3, 2])191 predict_runner(train_pattern, targets, predictions, module_sizes, mode='log')192 class TestDifferentTestPattern:193 def testModuleSize2_2_log(self):194 train_pattern = np.array([[1, 0], [0, 1]])195 targets = np.array([[1, 0], [0, 1]])196 test_pattern = np.array([[1/2, 1/2], [2/3, 1/3]])197 predictions = np.array([[np.log(5/8), np.log(5/8)], [np.log(3/4), np.log(1/2)]])198 module_sizes = np.array([2, 2])199 predict_runner(train_pattern, targets, predictions, module_sizes, test_pattern=test_pattern, mode='log')200 def testModuleSize2_2_predict(self):201 train_pattern = np.array([[1, 0], [0, 1]])202 targets = np.array([[1, 0], [0, 1]])203 test_pattern = np.array([[1/2, 1/2], [2/3, 1/3]])204 predictions = np.array([[np.log(5/8), np.log(5/8)], [np.log(3/4), np.log(1/2)]])205 predictions = np.array([0, 0])206 module_sizes = np.array([2, 2])207 predict_runner(train_pattern, targets, predictions, module_sizes, test_pattern=test_pattern, mode='predict')208 def testModuleSize2_2_single_log(self):209 train_pattern = np.array([[1, 0]])210 targets = np.array([[1, 0]])211 test_pattern = np.array([[1, 0], [2/3, 1/3]])212 predictions = np.array([[0, 0], [0, 0]])213 module_sizes = np.array([2, 2])214 predict_runner(train_pattern, targets, predictions, module_sizes, test_pattern=test_pattern, mode='log')215 def testModuleSize3_2_log(self):216 train_pattern = np.array([[1, 0, 0], [0, 1, 0]])217 targets = np.array([[1, 0], [0, 1]])218 test_pattern = np.array([[0, 0, 1], [0, 1/2, 1/2]])219 predictions = np.array([[np.log(1/2), np.log(1/2)], [np.log(3/8), np.log(3/4)]])220 module_sizes = np.array([3, 2])221 predict_runner(train_pattern, targets, predictions, module_sizes, test_pattern=test_pattern, mode='log')222 def testModuleSize3_2_predict(self):223 train_pattern = np.array([[1, 0, 0], [0, 1, 0]])224 targets = np.array([[1, 0], [0, 1]])225 test_pattern = np.array([[0, 0, 1], [0, 1/2, 1/2]])226 predictions = np.array([[np.log(1/2), np.log(1/2)], [np.log(3/8), np.log(3/4)]])227 predictions = np.array([0, 1])228 module_sizes = np.array([3, 2])...
regular_re.py
Source:regular_re.py
1import re2# pattern to search for3patterns = ['term1', 'term2', 'term3', 'term']4# string to search pattern in5text = 'This is a string with term1, and not the other term'6# for pattern in patterns:7# print('I am searching for: {}'.format(pattern)) # code to execute each time a match is found8# Using re module functionality to perform the search 9for pattern in patterns:10 print('I am searching for: {}'.format(pattern))11 if re.search(pattern, text):12 print('MATCH FOUND')13 else:14 print('NO MATCH!')15# re.search() returns special regular expressions object <class '_sre.SRE_Match'> which contains more than the booleanes of the search, it contains other details e.g where in the text the match was found e.g: at what index withn the string, using the start() method16match_info = re.search(pattern, text).start()17print(match_info)18# You can also split the text, returns a list of the split part of the string19split_term = '@'20email = 'user@email.com'21print(re.split(split_term, email))22# Using re to find all instance of a pattern, returning a list of of all non-overlapping matches23print(re.findall('match', 'This text has two match strings that need to find match for'))24# Using metacharacters to find repetition within25def multi_re_find(patterns, phrase):26 for pat in patterns:27 print('Searching for pattern: %s') %pattern28 print(re.findall(pat, phrase))29 print('\n') # print a new line after every search pass30# Finding patterns31test_phrase = 'sdsd..sssddd...sdddsddd...dsds...dsssss...sdddd'32test_pattern = ['sd*'] # returns a list with all patterns featuring 's' and 'd' repeated zero or more times33multi_re_find(test_pattern, test_phrase)34test_pattern = ['sd+'] # returns a list with all patterns featuring 's' and 'd' repeated one or more times35multi_re_find(test_pattern, test_phrase)36test_pattern = ['sd?'] # returns a list with all patterns featuring 's' and 'd' repeated zero or once37multi_re_find(test_pattern, test_phrase)38test_pattern = ['sd{3}'] # returns a list with all patterns featuring 's' and 'd' a specific number of times39multi_re_find(test_pattern, test_phrase)40test_pattern = ['sd{1,3}'] # returns a list with all patterns featuring 's' and 'd' a specific number of times or another specific number of times41multi_re_find(test_pattern, test_phrase)42test_pattern = ['s[sd]+'] # returns a list with all patterns featuring 's' and followed by either 1 or more 's' or 'd'.43multi_re_find(test_pattern, test_phrase)44# Exclusions using the 'carrot' ^ symbol45test_phrase = 'This is a string! But it has punctuation. How can we remove it?'46test_pattern = ['[^!.?]+'] # Splits the string where the punctuations appear one or more times47multi_re_find(test_pattern, test_phrase)48# Sequences of lower case letters: ['[a-z]+'], uppercase ['[A-Z]+'], both upper and lower case ['[a-zA-Z]+'], one upper followed by one or more lower case ['[A-Z][a-z]+'] etc49# Finding escape characters, which in python are prefixed by '\'. However the '\' must itself be escaped in a normal string, which is done by making a literal value using 'r'. 50test_phrase = 'This is a string890! But it has numbers 22. How can we #remove 12them?'51test_pattern = [r'\d+'] # returns sequences of one or more digits52multi_re_find(test_pattern, test_phrase)53test_phrase = 'This is a string890! But it has numbers 22. How can we #remove 12them?'54test_pattern = [r'\s+'] # returns sequences of one or more spaces55multi_re_find(test_pattern, test_phrase)56test_phrase = 'This is a string890! But it has numbers 22. How can we #remove 12them?'57test_pattern = [r'\D+'] # returns sequences of one or more non-digits58multi_re_find(test_pattern, test_phrase)59test_phrase = 'This is a string890! But it has numbers 22. How can we #remove 12them?'60test_pattern = [r'\S+'] # returns sequences of one or more non-spaces61multi_re_find(test_pattern, test_phrase)62test_phrase = 'This is a string890! But it has numbers 22. How can we #remove 12them?'63test_pattern = [r'\w+'] # returns sequences of one or more alphanumerics64multi_re_find(test_pattern, test_phrase)65test_phrase = 'This is a string890! But it has numbers 22. How can we #remove 12them?'66test_pattern = [r'\W+'] # returns sequences of one or more non-alphanumeric...
Automation Testing Tutorials
Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.
LambdaTest Learning Hubs:
- JUnit Tutorial
- TestNG Tutorial
- Webdriver Tutorial
- WebDriverIO Tutorial
- Protractor Tutorial
- Selenium 4 Tutorial
- Jenkins Tutorial
- NUnit Tutorial
- Jest Tutorial
- Playwright Tutorial
- Cypress Tutorial
- PyTest Tutorial
YouTube
You could also refer to video tutorials over LambdaTest YouTube channel to get step by step demonstration from industry experts.
Run lisa automation tests on LambdaTest cloud grid
Perform automation testing on 3000+ real desktop and mobile devices online.
Try LambdaTest Now !!
Get 100 minutes of automation test minutes FREE!!
Was this article helpful?
