How to use test_new method in pytest-django

Best Python code snippet using pytest-django_python

test_jama_sync_gherkin.py

Source:test_jama_sync_gherkin.py

1import jama_sync as jama_sync2import pytest3from pathlib import Path4from libraries.unit_test_helper import replace_section_in_file, standardize_file5testing = True6path_to_folder = str(Path(__file__).parent)7root_test_data_folder = path_to_folder + "/test_data/gherkin_jama_sync_test"8test_path = f"{root_test_data_folder}/test_docs.feature"9standard_file_path = path_to_folder + "/test_data/docs_standards/test_docs_standard.feature"10# Test that Jama Sync will accept correct documentation11def test_correct_documentation(capfd, username, password):12    arguments = jama_sync.parse_args(["-root_path", root_test_data_folder,13                                      "-test_path", test_path,14                                      "-jama_user", username, "-jama_pass", password,15                                      "-interpreter", "gherkin"])16    standardize_file(test_path, standard_file_path)17    jama_sync.update_jama(arguments, testing)18    output, err = capfd.readouterr()19    assert "the following 'gherkin' test cases were read:\n" \20           "test_NEW docs" in output21def test_add(capfd, username, password):22    arguments = jama_sync.parse_args(["-root_path", root_test_data_folder,23                                      "-test_path", test_path,24                                      "-jama_user", username, "-jama_pass", password,25                                      "-interpreter", "gherkin", "--add"])26    standardize_file(test_path, standard_file_path)27    jama_sync.update_jama(arguments, testing)28    output, err = capfd.readouterr()29    assert "the following 'gherkin' test cases were read:\n" \30           "test_NEW docs" in output31    assert "Successfully updated: test_gid_" in output32# WARNING: this test case depends on the last test case33@pytest.mark.dependency(depends=["test_add"])34def test_add_update(capfd, username, password):35    arguments = jama_sync.parse_args(["-root_path", root_test_data_folder,36                                      "-test_path", test_path,37                                      "-jama_user", username, "-jama_pass", password,38                                      "-interpreter", "gherkin", "--update"])39    jama_sync.update_jama(arguments, testing)40    output, err = capfd.readouterr()41    assert "the following 'gherkin' test cases were read:\ntest_" in output and "docs" in output42# WARNING: this test case depends on the last test case43@pytest.mark.dependency(depends=["test_add_update"])44@pytest.mark.parametrize("section, section_change",45                         [("Description:",46                           "        Testing Jama Sync table writing in Description\n"47                           "        *begin table*\n"48                           "        testing | testing | testing |\n"49                           "        testing | testing | testing |\n"50                           "        *end table*"),51                          ("Prerequisites:",52                           "        1) Testing Jama Sync table writing in Prerequisites\n"53                           "        *begin table*\n"54                           "        testing | testing | testing\n"55                           "        testing | testing | testing\n"56                           "        *end table*"),57                          ("Test Data:",58                           "        1) Testing Jama Sync table writing in Test Data\n"59                           "        *begin table*\n"60                           "        testing | testing | testing\n"61                           "        testing | testing | testing\n"62                           "        *end table*"),63                          ]64                         )65def test_table_documentation_happy_path(capfd, username, password, section, section_change):66    arguments = jama_sync.parse_args(["-root_path", root_test_data_folder,67                                      "-test_path", test_path,68                                      "-jama_user", username, "-jama_pass", password,69                                      "-interpreter", "gherkin", "--update"])70    replace_section_in_file(test_path, section, section_change, "gherkin")71    jama_sync.update_jama(arguments, testing)72    output, err = capfd.readouterr()73    assert "the following 'gherkin' test cases were read:\ntest_" in output and "docs" in output74@pytest.mark.parametrize("section, delete_section, expected_string",75                         [("Description:", True, "Expecting 'Description' line in test case test_NEW docs "76                                                 "(if this section is present, check that it is the first line "77                                                 "in the doc-string)"),78                          ("Description:", False, "Expecting 'Prerequisites' line in test case test_NEW docs or the "79                                                  "previous section data (if this section is present, check that it "80                                                  "is directly 2 lines below the last section)"),81                          ("Prerequisites:", True, "Expecting 'Prerequisites' line in test case test_NEW docs or the "82                                                   "previous section data (if this section is present, check that it "83                                                   "is directly 2 lines below the last section)"),84                          ("Prerequisites:", False, "Failed to read test file: Expecting either '1)', another dash "85                                                    "'-', or '*begin table*' in the Prerequisites section in "86                                                    "test_NEW docs"),87                          ("Test Data:", True, "Expecting 'Test Data' line in test case test_NEW docs or the "88                                               "previous section data (if this section is present, check that it "89                                               "is directly 2 lines below the last section)"),90                          ("Test Data:", False, "Failed to read test file: Expecting either '1)', another dash '-', or"91                                                " '*begin table*' in the Test Data section in test_NEW docs"),92                          ("Steps:", True, "Expecting 'Steps' line in test case test_NEW docs or the previous section "93                                           "data (if this section is present, check that it is directly 2 lines below "94                                           "the last section)"),95                          ("Steps:", False, "Expecting '1)' line under 'Steps' in test case test_NEW docs"),96                          ]97                         )98def test_basic_incorrect_documentation(capfd, username, password, section, delete_section, expected_string):99    arguments = jama_sync.parse_args(["-root_path", root_test_data_folder,100                                      "-test_path", test_path,101                                      "-jama_user", username, "-jama_pass", password,102                                      "-interpreter", "gherkin", "--add"])103    standardize_file(test_path, standard_file_path)104    replace_section_in_file(test_path, section, "", "gherkin", delete_section=delete_section)105    jama_sync.update_jama(arguments, testing)106    output, err = capfd.readouterr()107    if expected_string not in output:108        print(output)109    assert expected_string in output110@pytest.mark.parametrize("section_change, expected_string",111                         [("        1) Testing Jama Sync writing in Steps\n"112                           "            Notes: Testing Jama Sync writing in Notes",113                           "Expecting 'ER:' line under 'Steps' in test case test_NEW docs"),114                          ("        1) Testing Jama Sync writing in Steps\n"115                           "            ER: Testing Jama Sync writing in Expected Results",116                           "Expecting 'Notes:' line under 'Steps' in test case test_NEW docs"),117                          ]118                         )119def test_incorrect_steps_documentation(capfd, username, password, section_change, expected_string):120    arguments = jama_sync.parse_args(["-root_path", root_test_data_folder,121                                      "-test_path", test_path,122                                      "-jama_user", username, "-jama_pass", password,123                                      "-interpreter", "gherkin", "--add"])124    standardize_file(test_path, standard_file_path)125    replace_section_in_file(test_path, "Steps:", section_change, "gherkin")126    jama_sync.update_jama(arguments, testing)127    output, err = capfd.readouterr()128    if expected_string not in output:129        print(output)130    assert expected_string in output131@pytest.mark.parametrize("section, section_change, expected_string",132                         [("Prerequisites:",133                           "        1) Testing Jama Sync writing in Prerequisites\n"134                           "        3) Testing Jama Sync writing in Prerequisites\n",135                           "Expecting either '2)', another dash '-', or '*begin table*' in the "136                           "Prerequisites section in test_NEW docs"),137                          ("Test Data:",138                           "        1) Testing Jama Sync writing in Test Data\n"139                           "        3) Testing Jama Sync writing in Test Data\n",140                           "Expecting either '2)', another dash '-', or '*begin table*' in the "141                           "Test Data section in test_NEW docs"),142                          ("Steps:",143                           "        1) Testing Jama Sync writing in Steps\n"144                           "            ER: Testing Jama Sync writing in Expected Results\n"145                           "            Notes: Testing Jama Sync writing in Notes\n"146                           "        3) Testing Jama Sync writing in Steps\n",147                           "Expecting '2)' line under 'Steps' in test case test_NEW docs"),148                          ]149                         )150def test_incorrect_numbering_documentation(capfd, username, password, section, section_change, expected_string):151    arguments = jama_sync.parse_args(["-root_path", root_test_data_folder,152                                      "-test_path", test_path,153                                      "-jama_user", username, "-jama_pass", password,154                                      "-interpreter", "gherkin", "--add"])155    standardize_file(test_path, standard_file_path)156    replace_section_in_file(test_path, section, section_change, "gherkin")157    jama_sync.update_jama(arguments, testing)158    output, err = capfd.readouterr()159    if expected_string not in output:160        print(output)161    assert expected_string in output162@pytest.mark.parametrize("section, section_change, expected_string",163                         [("Description:",164                           "        Testing Jama Sync table writing in Description\n"165                           "        *begin table*\n"166                           "        testing | testing | testing\n"167                           "        testing | testing\n"168                           "        *end table*",169                           "Table row: 'testing | testing' does not have the expected amount of columns:3 as the rest "170                           "of the table in test_NEW docs"),171                          ("Description:",172                           "        Testing Jama Sync table writing in Description\n"173                           "        *begin table*\n"174                           "        testing | testing | testing\n"175                           "        testing | testing\n",176                           "Expecting '*end table*' in the Description section in test_NEW docs"),177                          ("Prerequisites:",178                           "        1) Testing Jama Sync table writing in Prerequisites\n"179                           "        *begin table*\n"180                           "        testing | testing | testing\n"181                           "        testing | testing\n"182                           "        *end table*",183                           "Table row: 'testing | testing' does not have the expected amount of columns:3 as the rest "184                           "of the table in test_NEW docs"),185                          ("Prerequisites:",186                           "        1) Testing Jama Sync table writing in Prerequisites\n"187                           "        *begin table*\n"188                           "        testing | testing | testing\n"189                           "        testing | testing\n",190                           "Expecting '*end table*' in the Prerequisites section in test_NEW docs"),191                          ("Test Data:",192                           "        1) Testing Jama Sync table writing in Test Data\n"193                           "        *begin table*\n"194                           "        testing | testing | testing\n"195                           "        testing | testing\n"196                           "        *end table*",197                           "Table row: 'testing | testing' does not have the expected amount of columns:3 as the rest "198                           "of the table in test_NEW docs"),199                          ("Test Data:",200                           "        1) Testing Jama Sync table writing in Test Data\n"201                           "        *begin table*\n"202                           "        testing | testing | testing\n"203                           "        testing | testing\n",204                           "Expecting '*end table*' in the Test Data section in test_NEW docs"),205                          ]206                         )207def test_incorrect_table_documentation(capfd, username, password, section, section_change, expected_string):208    arguments = jama_sync.parse_args(["-root_path", root_test_data_folder,209                                      "-test_path", test_path,210                                      "-jama_user", username, "-jama_pass", password,211                                      "-interpreter", "gherkin", "--add"])212    standardize_file(test_path, standard_file_path)213    replace_section_in_file(test_path, section, section_change, "gherkin")214    jama_sync.update_jama(arguments, testing)215    output, err = capfd.readouterr()216    if expected_string not in output:217        print(output)...

eda.py

Source:eda.py

1import pandas as pd2from sklearn import preprocessing3import numpy as np4from eda_tools import *5PATH = "D:/Documents/Machine Learning/Walmart Trip Type"6DATA_IMPORT_PATH = PATH + "/data/"78CAT_DUMMIES = True9DAY_DUMMIES = True10DAY_SIN_COS = True11DPT_ENCODE = True12FAVOURITE_FINELINES =True13fav_fnls = 414PERC_NAN = 015export = True1617if __name__ == '__main__':1819    train, test, sample_submission = read_data()2021    # enumareate days22    train = enum_days(train)23    test = enum_days(test)2425    # and calculate day index26    train = day_index(train)27    test = day_index(test)2829    # calculate week number30    train, test = week_number(train, test)31    test_correct = no_dup_upcs(test)32    train_correct = no_dup_upcs(train)3334    #  visitnumber scancount sum, total items purchased35    train_new = train.groupby("VisitNumber")["ScanCount"]\36                     .agg("sum")\37                     .reset_index()\38                     .rename(columns={"ScanCount": "CartNetItems"})39    test_new = test.groupby("VisitNumber")["ScanCount"]\40                   .agg("sum")\41                   .reset_index()\42                   .rename(columns={"ScanCount": "CartNetItems"})43    train_new = items_bought(train, train_new)44    test_new = items_bought(test, test_new)4546    # add triptype to each visitnumber47    train_new = add_triptype(train, train_new)4849    # rearrange columns50    train_new = train_new[["VisitNumber", "TripType", "CartNetItems", "ItemsBought"]]5152    # fineline entropy53    train_new = fineline_entropy(train, train_new)54    test_new = fineline_entropy(test, test_new)55    #fineline chaos56    train_new = fineline_chaos(train, train_new)57    test_new = fineline_chaos(test, test_new)5859    train_new = refunded_exact_items(train_correct, train_new)60    test_new = refunded_exact_items(test_correct, test_new)61    train_new = replaced_refunded_not_exact(train_correct, train_new)62    test_new = replaced_refunded_not_exact(test_correct, test_new)6364    # bought exactly the number of items i returned(same upc)65    train_new = replaced_exact_items(train_correct, train_new)66    test_new = replaced_exact_items(test_correct, test_new)6768    # total items returned69    train_new = items_returned(train, train_new)70    test_new = items_returned(test, test_new)7172    # returned to bought ratio73    train_new["RtnTobtRatio"] = (train_new["ItemsReturned"] / train_new["ItemsBought"]).round(3)74    test_new["RtnTobtRatio"] = (test_new["ItemsReturned"] / test_new["ItemsBought"]).round(3)7576    # Total Items77    train_new["TotalItems"] = train_new["ItemsBought"] + train_new["ItemsReturned"]78    test_new["TotalItems"] = test_new["ItemsBought"] + test_new["ItemsReturned"]7980    # add day eum81    temp = train.groupby(["VisitNumber", "WeekdayEnum"]).first().reset_index()82    train_new = train_new.merge(temp[["WeekdayEnum", "VisitNumber"]], on="VisitNumber", how="left")83    temp = test.groupby(["VisitNumber", "WeekdayEnum"]).first().reset_index()84    test_new = test_new.merge(temp[["WeekdayEnum", "VisitNumber"]], on="VisitNumber", how="left")8586    train["DepartmentDescription"] = train["DepartmentDescription"].fillna("None")87    test["DepartmentDescription"] = test["DepartmentDescription"].fillna("None")8889    # nunique UPCs for buyght items and returned90    temp = train[train["ScanCount"] > 0].groupby("VisitNumber")["Upc"].nunique().reset_index()91    train_new = train_new.merge(temp, how="left", on="VisitNumber").rename(columns={"Upc": "UniqueUpcsBought"})92    train_new["UniqueUpcsBought"] = train_new["UniqueUpcsBought"].fillna(0)9394    temp = test[test["ScanCount"] > 0].groupby("VisitNumber")["Upc"].nunique().reset_index()95    test_new = test_new.merge(temp, how="left", on="VisitNumber").rename(columns={"Upc": "UniqueUpcsBought"})96    test_new["UniqueUpcsBought"] = test_new["UniqueUpcsBought"].fillna(0)9798    temp = train[train["ScanCount"] < 0].groupby("VisitNumber")["Upc"].nunique().reset_index()99    train_new = train_new.merge(temp, how="left", on="VisitNumber").rename(columns={"Upc": "UniqueUpcsReturned"})100    train_new["UniqueUpcsReturned"] = train_new["UniqueUpcsReturned"].fillna(0)101102    temp = test[test["ScanCount"] < 0].groupby("VisitNumber")["Upc"].nunique().reset_index()103    test_new = test_new.merge(temp, how="left", on="VisitNumber").rename(columns={"Upc": "UniqueUpcsReturned"})104    test_new["UniqueUpcsReturned"] = test_new["UniqueUpcsReturned"].fillna(0)105106    # total unique UPCs107    train_new["TotalUniqueUpcs"] = train_new["UniqueUpcsBought"] + train_new["UniqueUpcsReturned"]108    test_new["TotalUniqueUpcs"] = test_new["UniqueUpcsBought"] + test_new["UniqueUpcsReturned"]109110    # total unique UPCs to total items perc111    train_new["TotalUniqueUpcs_perc"] = (train_new["TotalUniqueUpcs"] / train_new["TotalItems"]).round(3)112    test_new["TotalUniqueUpcs_perc"] = (test_new["TotalUniqueUpcs"] / test_new["TotalItems"]).round(3)113114    # UniqueUpcs bought /items bought115    train_new["UniqueUpcsBought_perc"] = (train_new["UniqueUpcsBought"] / train_new["ItemsBought"]).round(3)116    train_new["UniqueUpcsReturned_perc"] = (train_new["UniqueUpcsReturned"] / train_new["ItemsReturned"]).round(3)117118    train_new["UniqueUpcsBought_perc"] = train_new["UniqueUpcsBought_perc"].fillna(PERC_NAN)119    train_new["UniqueUpcsReturned_perc"] = train_new["UniqueUpcsReturned_perc"].fillna(PERC_NAN)120    test_new["UniqueUpcsBought_perc"] = (test_new["UniqueUpcsBought"] / test_new["ItemsBought"]).round(3)121    test_new["UniqueUpcsReturned_perc"] = (test_new["UniqueUpcsReturned"] / test_new["ItemsReturned"]).round(3)122123    test_new["UniqueUpcsBought_perc"] = test_new["UniqueUpcsBought_perc"].fillna(PERC_NAN)124    test_new["UniqueUpcsReturned_perc"] = test_new["UniqueUpcsReturned_perc"].fillna(PERC_NAN)125126    # unique finelineNumbers127    temp = train[train["ScanCount"] > 0].groupby("VisitNumber")["FinelineNumber"].nunique().reset_index()128    train_new = train_new.merge(temp, how="left", on="VisitNumber")\129                         .rename(columns={"FinelineNumber": "UniqueFinelinesBought"})130    train_new["UniqueFinelinesBought"] = train_new["UniqueFinelinesBought"].fillna(0)131132    temp = test[test["ScanCount"] > 0].groupby("VisitNumber")["FinelineNumber"].nunique().reset_index()133    test_new = test_new.merge(temp, how="left", on="VisitNumber")\134                       .rename(columns={"FinelineNumber": "UniqueFinelinesBought"})135136    test_new["UniqueFinelinesBought"] = test_new["UniqueFinelinesBought"].fillna(0)137138    temp = train[train["ScanCount"] < 0].groupby("VisitNumber")["FinelineNumber"]\139                                        .nunique()\140                                        .reset_index()141    train_new = train_new.merge(temp, how="left", on="VisitNumber")\142                         .rename(columns={"FinelineNumber": "UniqueFinelinesReturned"})143144    train_new["UniqueFinelinesReturned"] = train_new["UniqueFinelinesReturned"].fillna(0)145    temp = test[test["ScanCount"] < 0].groupby("VisitNumber")["FinelineNumber"]\146                                      .nunique()\147                                      .reset_index()148    test_new = test_new.merge(temp, how="left", on="VisitNumber")\149                       .rename(columns={"FinelineNumber": "UniqueFinelinesReturned"})150    test_new["UniqueFinelinesReturned"] = test_new["UniqueFinelinesReturned"].fillna(0)151152    # total Unique finelines153    train_new["TotalUniqueFinelines"] = train_new["UniqueFinelinesReturned"] + train_new["UniqueFinelinesBought"]154    test_new["TotalUniqueFinelines"] = test_new["UniqueFinelinesReturned"] + test_new["UniqueFinelinesBought"]155156    # total unique finlines to total items percentage157    train_new["TotalUniqueFinelines_perc"] = (train_new["TotalUniqueFinelines"] / train_new["TotalItems"]).round(3)158    test_new["TotalUniqueFinelines_perc"] = (test_new["TotalUniqueFinelines"] / test_new["TotalItems"]).round(3)159160    # finelines percentage161    train_new["UniqueFinelinesBought_perc"] = (train_new["UniqueFinelinesBought"] / train_new["ItemsBought"]).round(3)162    train_new["UniqueFinelinesReturned_perc"] = (train_new["UniqueFinelinesReturned"] / train_new["ItemsReturned"])\163                                                .round(3)164165    train_new["UniqueFinelinesBought_perc"] = train_new["UniqueFinelinesBought_perc"].fillna(PERC_NAN)166    train_new["UniqueFinelinesReturned_perc"] = train_new["UniqueFinelinesReturned_perc"].fillna(PERC_NAN)167    test_new["UniqueFinelinesBought_perc"] = (test_new["UniqueFinelinesBought"] / test_new["ItemsBought"]).round(3)168    test_new["UniqueFinelinesReturned_perc"] = (test_new["UniqueFinelinesReturned"] / test_new["ItemsReturned"]).round(169        3)170171    test_new["UniqueFinelinesBought_perc"] = test_new["UniqueFinelinesBought_perc"].fillna(PERC_NAN)172    test_new["UniqueFinelinesReturned_perc"] = test_new["UniqueFinelinesReturned_perc"].fillna(PERC_NAN)173174    # finelines entropy175176    train_new.replace(np.inf, 0, inplace=True)177    test_new.replace(np.inf, 0, inplace=True)178179    if FAVOURITE_FINELINES:180        train_new = get_fav_flnumbers(train, train_new, n=fav_fnls)181        test_new = get_fav_flnumbers(test, test_new, n=fav_fnls)182183    if CAT_DUMMIES:184        train_new = category_dummies(train, train_new, train=True)185        test_new = category_dummies(test, test_new)186        train_new = train_new.drop(columns="HEALTH AND BEAUTY AIDS", axis=1)187188    # onehot encode days189    if DAY_DUMMIES:190        train_new = day_dummies(train, train_new, train=True)191        test_new = day_dummies(test, test_new)192193    if DAY_SIN_COS:194        train_new = days_sin_cos(train_new)195        test_new = days_sin_cos(test_new)196        print()197    # encode department description198    if DPT_ENCODE:199        department_description_le = preprocessing.LabelEncoder()200        department_description_le.fit(pd.concat([train['DepartmentDescription'], test["DepartmentDescription"]]))201        department_description_mapping = dict(zip(department_description_le.classes_,202                                                  department_description_le.transform(department_description_le.classes_)))203        department_description_mapping_inv = dict(204            zip(department_description_le.transform(department_description_le.classes_),205                department_description_le.classes_))206        train["DepartmentDescription"] = train["DepartmentDescription"].map(department_description_mapping)207208    if export:209        chonda = ""210        if DAY_SIN_COS:211            chonda = chonda + "_sincos"212        if CAT_DUMMIES:213            chonda = chonda + "_catdum"214        if DAY_DUMMIES:215            chonda = chonda + "_daydum"216        if FAVOURITE_FINELINES:217            chonda = chonda + "_" + str(fav_fnls) + "favFL"218        train_new.to_csv(PATH + "/eda final/" + "train_new_temp" + chonda+".csv", index=False)219        # train.to_csv(PATH + "/temp/" + "train_temp.csv", index=False)220        # test.to_csv(PATH + "/temp/" + "test_temp.csv", index=False)221        test_new.to_csv(PATH + "/eda final/" + "test_new_temp" + chonda + ".csv", index=False)222
...

utils.py

Source:utils.py

1"""2Helper functions for categorical encodings3"""4from pandas.api.types import is_string_dtype, is_numeric_dtype5from sklearn.preprocessing import OneHotEncoder6from sklearn.feature_extraction import FeatureHasher7from sklearn.model_selection import KFold8import pandas as pd9from sklearn.ensemble import RandomForestClassifier10from sklearn.metrics import roc_auc_score11def kfold_target_encoder(train, test, cols_encode, target, folds=10):12    """13    Mean regularized target encoding based on kfold14    """15    train_new = train.copy()16    test_new = test.copy()17    kf = KFold(n_splits=folds, random_state=1)18    for col in cols_encode:19        global_mean = train_new[target].mean()20        for train_index, test_index in kf.split(train):21            mean_target = train_new.iloc[train_index].groupby(col)[target].mean()22            train_new.loc[test_index, col + "_mean_enc"] = train_new.loc[test_index, col].map(mean_target)23        train_new[col + "_mean_enc"].fillna(global_mean, inplace=True)24        # making test encoding using full training data25        col_mean = train_new.groupby(col)[target].mean()26        test_new[col + "_mean_enc"] = test_new[col].map(col_mean)27        test_new[col + "_mean_enc"].fillna(global_mean, inplace=True)28    29    # filtering only mean enc cols30    train_new = train_new.filter(like="mean_enc", axis=1)31    test_new = test_new.filter(like="mean_enc", axis=1)32    return train_new, test_new33        34def catboost_target_encoder(train, test, cols_encode, target):35    """36    Encoding based on ordering principle37    """38    train_new = train.copy()39    test_new = test.copy()40    for column in cols_encode:41        global_mean = train[target].mean()42        cumulative_sum = train.groupby(column)[target].cumsum() - train[target]43        cumulative_count = train.groupby(column).cumcount()44        train_new[column + "_cat_mean_enc"] = cumulative_sum/cumulative_count45        train_new[column + "_cat_mean_enc"].fillna(global_mean, inplace=True)46        # making test encoding using full training data47        col_mean = train_new.groupby(column).mean()[column + "_cat_mean_enc"]  #48        test_new[column + "_cat_mean_enc"] = test[column].map(col_mean)49        test_new[column + "_cat_mean_enc"].fillna(global_mean, inplace=True)50    # filtering only mean enc cols51    train_new = train_new.filter(like="cat_mean_enc", axis=1)52    test_new = test_new.filter(like="cat_mean_enc", axis=1)53    return train_new, test_new54def one_hot_encoder(train, test, cols_encode, target=None):55    """ one hot encoding"""56    ohc_enc = OneHotEncoder(handle_unknown='ignore')57    ohc_enc.fit(train[cols_encode])58    train_ohc = ohc_enc.transform(train[cols_encode])59    test_ohc = ohc_enc.transform(test[cols_encode])60    return train_ohc, test_ohc61    62def label_encoder(train, test, cols_encode=None, target=None):63    """64    Code borrowed from fast.ai and is tweaked a little.65    Convert columns in a training and test dataframe into numeric labels 66    """67    train_new = train.drop(target, axis=1).copy()68    test_new = test.drop(target, axis=1).copy()69    70    for n,c in train_new.items():71        if is_string_dtype(c) or n in cols_encode : train_new[n] = c.astype('category').cat.as_ordered()72    73    if test_new is not None:74        for n,c in test_new.items():75            if (n in train_new.columns) and (train_new[n].dtype.name=='category'):76                test_new[n] = pd.Categorical(c, categories=train_new[n].cat.categories, ordered=True)77            78    cols = list(train_new.columns[train_new.dtypes == 'category'])79    for c in cols:80        train_new[c] = train_new[c].astype('category').cat.codes81        if test_new is not None: test_new[c] = test_new[c].astype('category').cat.codes82    return train_new, test_new83def hash_encoder(train, test, cols_encode, target=None, n_features=10):84    """hash encoder"""85    h = FeatureHasher(n_features=n_features, input_type="string")86    for col_encode in cols_encode:87        h.fit(train[col_encode])88        train_hash = h.transform(train[col_encode])89        test_hash = h.transform(test[col_encode])90    return train_hash, test_hash91def fitmodel_and_auc_score(encoder, train, test, cols_encode, target, **kwargs):92    """93    Fits and returns scores of a random forest model. Uses ROCAUC as scoring metric94    """95    model = RandomForestClassifier(n_estimators=500,96                                   n_jobs=-1, 97                                   class_weight="balanced",98                                   max_depth=10)99    if encoder:100        train_encoder, test_encoder = encoder(train, test, cols_encode=cols_encode, target=target)101    else:102        train_encoder, test_encoder = train.drop(target, axis=1), test.drop(target, axis=1)103    model.fit(train_encoder, train[target])104    train_score = roc_auc_score(train[target], model.predict(train_encoder))105    valid_score = roc_auc_score(test[target], model.predict(test_encoder))...

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.