KaneAI - World’s First E2E Software Testing Agent.Book a Demo Join TestMu Conference

How to use test_pre_post method in avocado

Best Python code snippet using avocado_python

predictions.py

Source:predictions.py

1from sklearn.metrics import confusion_matrix, classification_report2from main_func.BertTraining import batch_size3from aux_func.data_preprocess import Preprocessor4import numpy as np5from aux_func.confussion_matrix import plot_confusion_matrix6import tensorflow as tf7import os8# print(tf.version)9from aux_func.load_model import load_model10gpus = tf.config.experimental.list_physical_devices('GPU')11if gpus:12    try:13        # Currently, memory growth needs to be the same across GPUs14        for gpu in gpus:15            tf.config.experimental.set_memory_growth(gpu, True)16    except RuntimeError as e:17        # Memory growth must be set before GPUs have been initialized18        print(e)19def test_model(model_path, conjunto, patient, prueba=1, combination='mean', mode=0):20    """21    FunciÃ³n que prueba el modelo especificado.22    Esta funcion genera los reportes de clasificacion y matrices de confusion tanto para23    los eegs completos como para los fragmentos. Ademas si el modelo es por zonas tambien genera24    los valores para estas.25    :param model_path: ruta del modelo (La carpeta donde se almacena el modelo)26    :param conjunto: nÃºmero del 0 al 4: 'test', 'train', 'val', 'full', 'test_pre_post'27    :param patient: nÃºmero del -1 al 2: All, control, pre, post28    :param prueba: prueba del dataset:29        -1: "Both",30        0: "FTD",31        1: "FTI",32        2: "Resting"33    :param combination: 'mean' o 'majority_voting' para la combinaciÃ³n de los resultados en el34        modelo de zonas35    :param mode: 0: full y chunks, 1 full, 2 chunks36    :return: nada37    """38    # out_shape = [window_width, 64]39    model = load_model(model_path)40    conjuntos = ['test', 'train', 'val', 'full', 'test_pre_post']41    patients = ['control', 'pre', 'post', 'Pre-Post']42    pruebas = {-1: "Both",43               0: "FTD",44               1: "FTI",45               2: "Resting"}46    base_path = f'{conjuntos[conjunto]}_{patients[patient]}_{pruebas[prueba]}'47    zones = False48    if 'Zone' in model_path:49        base_path = f'{base_path}_{combination}'50        zones = True51    print(f'{base_path}')52    channels = []53    if model.input_shape[2] == 25:54        channels = [9, 10, 11, 12, 13, 19, 20, 21, 22, 23, 29, 30, 31, 32, 33, 39, 40, 41, 42, 43, 49, 50, 51, 52, 53]55    out_shape = list(model.input_shape[1:])56    test_post = False57    if conjunto == 4:58        test_post = True59    prepro = Preprocessor(batch_size,60                          model.input_shape[1],61                          64,62                          prueba=prueba,63                          limpio=0,64                          paciente=patient,65                          channels=channels,66                          transpose=True,67                          output_shape=out_shape,68                          test_post=test_post,69                          shuffle=False)70    try:71        os.mkdir(f'{model_path}/{base_path}')72    except Exception:73        pass74    if mode != 2:75        try:76            os.mkdir(f'{model_path}/{base_path}/full_eeg')77        except Exception:78            pass79        if conjunto == 0 or conjunto == 4:80            data = prepro.test_set81        if conjunto == 1:82            data = prepro.train_set83        if conjunto == 2:84            data = prepro.val_set85        if conjunto == 3:86            data = prepro.dataset87        y_pred = []88        y_pred_zones = [[] for _ in range(8)]89        for x_data, y_data in zip(data[0], data[1]):90            test_dataset = prepro.tf_from_generator([x_data], [y_data])91            pred = model.predict(test_dataset, verbose=1)92            if zones:93                pred, zone_pred = pred94                for i, zone in enumerate(np.swapaxes(zone_pred, 0, 1)):95                    y_pred_zones[i].append(np.argmax(np.asarray(np.mean(zone, axis=0))))96                    try:97                        os.mkdir(f'{model_path}/{base_path}/chunks/zones')98                        os.mkdir(f'{model_path}/{base_path}/full_eeg/zones')99                    except Exception:100                        pass101                if combination == 'majority_voting':102                    max_pred_total = np.bincount(np.array(y_pred_zones)[:, -1]).argmax()103                else:104                    max_pred_total = np.mean(np.array(y_pred_zones), axis=1).argmax()105                y_pred.append(max_pred_total)106            else:107                y_pred.append(np.mean(pred, axis=0))108                # FULL EEGS FIRST109        if not zones or combination == 'mean':110            y_pred = np.argmax(np.asarray(y_pred), axis=1)111        cf_matrix = confusion_matrix(data[1], y_pred)112        with open(f'{model_path}/{base_path}/full_eeg/classification_report.txt',113                  'w') as f:114            print(classification_report(data[1], y_pred, labels=[0, 1], target_names=["No Parkinson", "Parkinson"]), file=f)115        print(cf_matrix)116        plot_confusion_matrix(cm=cf_matrix,117                              normalize=False,118                              target_names=["No Parkinson", "Parkinson"],119                              title="Matriz de confusiÃ³n",120                              save=f'{model_path}/{base_path}/full_eeg/test_eeg.png')121        if y_pred_zones[0]:122            for i, y_pred in enumerate(y_pred_zones):123                cf_matrix = confusion_matrix(data[1], y_pred)124                print(cf_matrix)125                plot_confusion_matrix(cm=cf_matrix,126                                      normalize=False,127                                      target_names=["No Parkinson", "Parkinson"],128                                      title="Matriz de confusiÃ³n",129                                      save=f'{model_path}/{base_path}/full_eeg/zones/test_confussion_zone_{i + 1}_matrix.png')130    # CHUNKS NOW131    if mode != 1:132        try:133            os.mkdir(f'{model_path}/{base_path}/chunks')134        except Exception:135            pass136        full, train, test, val = prepro.classification_generator_dataset()137        dataset, train_dataset, test_dataset, val_dataset = prepro.classification_tensorflow_dataset()138        if conjunto == 0 or conjunto == 4:139            _, y_data = zip(*list(test))140            data_dataset = test_dataset141        if conjunto == 1:142            _, y_data = zip(*list(train))143            data_dataset = train_dataset144        if conjunto == 2:145            _, y_data = zip(*list(val))146            data_dataset = val_dataset147        if conjunto == 3:148            _, y_data = zip(*list(full))149            data_dataset = dataset150        y_data = list(y_data)151        print("Chunks")152        y_pred = model.predict(data_dataset, verbose=1)153        y_pred_zones = []154        if len(y_pred) == 2:155            y_pred, y_pred_zones = y_pred156            y_pred_zones = np.argmax(np.swapaxes(y_pred_zones, 0, 1), axis=2)157        y_pred = np.argmax(y_pred, axis=1)158        cf_matrix = confusion_matrix(y_data, y_pred)159        with open(f'{model_path}/{base_path}/chunks/classification_report.txt',160                  'w') as f:161            print(classification_report(y_data, y_pred, labels=[0, 1], target_names=["No Parkinson", "Parkinson"]), file=f)162        print(cf_matrix)163        plot_confusion_matrix(cm=cf_matrix,164                              normalize=False,165                              target_names=["No Parkinson", "Parkinson"],166                              title="Matriz de confusiÃ³n",167                              save=f'{model_path}/{base_path}/chunks/test_eeg.png')168        if y_pred_zones != []:169            for i, y_pred in enumerate(y_pred_zones):170                cf_matrix = confusion_matrix(y_data, y_pred)171                print(cf_matrix)172                plot_confusion_matrix(cm=cf_matrix,173                                      normalize=False,174                                      target_names=["No Parkinson", "Parkinson"],175                                      title="Matriz de confusiÃ³n",176                                      save=f'{model_path}/{base_path}/chunks/zones/test_confussion_zone_{i + 1}_matrix.png')177if __name__ == "__main__":178    model_paths = [#"C:/Users/Ceiec01/OneDrive - UFV/PFG/Codigo/checkpoints/BERT-ControlesPre-CanalesReducidos",179                   #"C:/Users/Ceiec01/OneDrive - UFV/PFG/Codigo/checkpoints/BERT-HigherDropout-64",180                   "C:/Users/Ceiec01/OneDrive - UFV/PFG/Codigo/checkpoints/BERT-Zones-Final"181                   ]182    for model_path in model_paths:183        for conjunto in range(3):184            print(f'{model_path} - {conjunto}')185            test_model(model_path, conjunto, 1)186        print(f'{model_path} - post')...

test_simple.py

Source:test_simple.py

...8    s = pd.Series([1, 2, 3, 4, 5], index=[1, 2, 3, 4, 5])9    obtained = s.context(3, around=2)10    expected = pd.Series([1, 2, 3, 4, 5], index=[1, 2, 3, 4, 5])11    pd.testing.assert_series_equal(obtained, expected)12def test_pre_post():13    s = pd.Series([1, 2, 3, 4, 5], index=[1, 2, 3, 4, 5])14    obtained = s.context(3, pre=2, post=1)15    expected = pd.Series([1, 2, 3, 4], index=[1, 2, 3, 4])...

test_tn.py

Source:test_tn.py

1from lcpy import TreeNode, build_root2def test_pre_post():3    l = [1,2,3,4,5,6,7]4    root = build_root(l)5    # print(root.post)6    assert root.pre == [1,2,4,5,3,6,7]...

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.