Join the 1-day Testing & QA Summit featuring 15+ Expert Speakers.Register for FREE! Join TestMu Conference

How to use test_sets method in assertpy

Best Python code snippet using assertpy_python

linreg.py

Source:linreg.py

1import csv2import numpy as np3import matplotlib.pyplot as plt4def read_data(filename):5	data = []6	with open(filename) as csvfile:7		reader = csv.reader(csvfile, quoting=csv.QUOTE_NONNUMERIC)8		for row in reader:9			data.append(row)10	return np.array(data)11def mse(l, train, test, train_regs, test_regs):12	tp = np.matrix.transpose(train)13	i = np.identity(len(train[0]))14	train_t = np.matrix.transpose(train_regs)[0]15	test_t = np.matrix.transpose(test_regs)[0]16	w = np.dot(np.matmul(np.linalg.inv((l * i) + np.matmul(tp, train)), tp), train_t)17	return np.mean(np.square(np.matmul(test, w) - test_t))	18def task1(train_sets, train_regs, test_sets, test_regs):19	sets = ["100-10", "100-100", "1000-100", "forest_fire", "real_estate", "50(1000)-100", "100(1000)-100", "150(1000)-100"]20	print("Lambda\tMSE")21	for i in range(len(train_sets)):22		train_mse = []23		test_mse = []24		for l in range(0, 151):25			train_mse.append(mse(l, train_sets[i], train_sets[i], train_regs[i], train_regs[i]))26			test_mse.append(mse(l, train_sets[i], test_sets[i], train_regs[i], test_regs[i]))27		plt.plot(range(0, 151), train_mse, label = "Train")28		plt.plot(range(0, 151), test_mse, label = "Test")29		plt.legend()30		plt.ylabel("MSE")31		plt.xlabel("Lambda")32		plt.title(sets[i])33		plt.show()34		print(str(np.argmin(test_mse)) + "\t" + str(min(test_mse)))35def task2(train_sets, train_regs, test_sets, test_regs):36	print("Lambda\tMSE")37	for i in range(len(train_sets)):38		mse_folds = []39		for j in range (0, 10):40			test_nums = range(0, len(train_sets[i]))[j::10]41			train_nums = np.setdiff1d(range(0, len(train_sets[i])), test_nums)42			train = np.take(train_sets[i], train_nums, axis = 0)43			test = np.take(train_sets[i], test_nums, axis = 0)44			train_reg = np.take(train_regs[i], train_nums)45			test_reg = np.take(train_regs[i], test_nums)46			47			mses = []48			for l in range(0, 151):49				mses.append(mse(l, train, test, train_reg, test_reg))50			mse_folds.append(mses)51		opt_l = np.argmin(np.sum(mse_folds, axis = 0))52		test_mse = mse(l, train_sets[i], test_sets[i], train_regs[i], test_regs[i])53		print(str(opt_l) + "\t" + str(test_mse))54	return55def task3(train_sets, train_regs, test_sets, test_regs):56	print("Alpha\tBeta\tMSE")57	for i in range(0, 8):58		train = train_sets[i]59		tp = np.matrix.transpose(train)60		r = train_regs[i]61		l = np.linalg.eigvals(np.matmul(tp, train))62		m = 0	63		a = 164		b = 165		diff = 166		while (diff > .0000000001):67			s = np.linalg.inv(a * np.identity(len(train[0])) + b * np.matmul(tp, train))68			m = np.matrix.transpose(b * np.dot(np.matmul(s, tp), r))[0]69			c = np.sum(np.divide(b * l, (a + b * l)))70			newa = c / np.matmul(np.matrix.transpose(m), m)71			newb = 1.0/((1.0/(len(train) - c)) * np.sum(np.square(r - np.dot(m, tp))))72			diff = min(abs(a - newa), abs(b - newb))73			a = newa74			b = newb75		test_mse = np.mean(np.square(np.matmul(test_sets[i], m) - test_regs[i]))76		print(str(np.real(a)) + "\t" + str(np.real(b)) + "\t" + str(np.real(test_mse)))77	return78	79if __name__ == "__main__":80	train_sets = []81	train_sets.append(read_data("train-100-10.csv"))82	train_sets.append(read_data("train-100-100.csv"))	83	train_sets.append(read_data("train-1000-100.csv"))84	train_sets.append(read_data("train-forestfire.csv"))85	train_sets.append(read_data("train-realestate.csv"))86	train_sets.append(train_sets[2][:50])87	train_sets.append(train_sets[2][:100])88	train_sets.append(train_sets[2][:150])89	train_regs = []90	train_regs.append(read_data("trainR-100-10.csv"))91        train_regs.append(read_data("trainR-100-100.csv"))92        train_regs.append(read_data("trainR-1000-100.csv"))93        train_regs.append(read_data("trainR-forestfire.csv"))94        train_regs.append(read_data("trainR-realestate.csv"))95        train_regs.append(train_regs[2][:50])96        train_regs.append(train_regs[2][:100])97        train_regs.append(train_regs[2][:150])98	99	test_sets = []100	test_sets.append(read_data("test-100-10.csv"))101        test_sets.append(read_data("test-100-100.csv"))102        test_sets.append(read_data("test-1000-100.csv"))103        test_sets.append(read_data("test-forestfire.csv"))104        test_sets.append(read_data("test-realestate.csv"))105	test_sets.append(test_sets[2])106	test_sets.append(test_sets[2])107	test_sets.append(test_sets[2])108	test_regs = []109	test_regs.append(read_data("testR-100-10.csv"))110        test_regs.append(read_data("testR-100-100.csv"))111        test_regs.append(read_data("testR-1000-100.csv"))112        test_regs.append(read_data("testR-forestfire.csv"))113        test_regs.append(read_data("testR-realestate.csv"))114        test_regs.append(test_regs[2])115        test_regs.append(test_regs[2])116        test_regs.append(test_regs[2])117	task1(train_sets, train_regs, test_sets, test_regs)118	task2(train_sets, train_regs, test_sets, test_regs)...

constants.py

Source:constants.py

1from pathlib import Path2# Paths3PACKAGE_DIR = Path(__file__).resolve().parent.parent4RESOURCES_DIR = PACKAGE_DIR / "resources"5TOOLS_DIR = RESOURCES_DIR / "tools"6DATA_DIR = RESOURCES_DIR / "data"7STANFORD_CORENLP_DIR = TOOLS_DIR / "stanford-corenlp-full-2018-10-05"8UCCA_DIR = TOOLS_DIR / "ucca-bilstm-1.3.10"9UCCA_PARSER_PATH = UCCA_DIR / "models/ucca-bilstm"10TEST_SETS_PATHS = {11    ('asset_test', 'orig'): DATA_DIR / f'test_sets/asset/asset.test.orig',12    ('asset_test', 'refs'): [DATA_DIR / f'test_sets/asset/asset.test.simp.{i}' for i in range(10)],13    ('asset_valid', 'orig'): DATA_DIR / f'test_sets/asset/asset.valid.orig',14    ('asset_valid', 'refs'): [DATA_DIR / f'test_sets/asset/asset.valid.simp.{i}' for i in range(10)],15    ('turkcorpus_test', 'orig'): DATA_DIR / f'test_sets/turkcorpus/test.truecase.detok.orig',16    ('turkcorpus_test', 'refs'): [DATA_DIR / f'test_sets/turkcorpus/test.truecase.detok.simp.{i}' for i in range(8)],17    ('turkcorpus_valid', 'orig'): DATA_DIR / f'test_sets/turkcorpus/tune.truecase.detok.orig',18    ('turkcorpus_valid', 'refs'): [DATA_DIR / f'test_sets/turkcorpus/tune.truecase.detok.simp.{i}' for i in range(8)],19    ('turkcorpus_test_legacy', 'orig'): DATA_DIR / f'test_sets/turkcorpus/legacy/test.8turkers.tok.norm',20    ('turkcorpus_test_legacy', 'refs'): [21        DATA_DIR / f'test_sets/turkcorpus/legacy/test.8turkers.tok.turk.{i}' for i in range(8)22    ],23    ('turkcorpus_valid_legacy', 'orig'): DATA_DIR / f'test_sets/turkcorpus/legacy/tune.8turkers.tok.norm',24    ('turkcorpus_valid_legacy', 'refs'): [25        DATA_DIR / f'test_sets/turkcorpus/legacy/tune.8turkers.tok.turk.{i}' for i in range(8)26    ],27    ('pwkp_test', 'orig'): DATA_DIR / f'test_sets/pwkp/pwkp.test.orig',28    ('pwkp_test', 'refs'): [DATA_DIR / f'test_sets/pwkp/pwkp.test.simp'],29    ('pwkp_valid', 'orig'): DATA_DIR / f'test_sets/pwkp/pwkp.valid.orig',30    ('pwkp_valid', 'refs'): [DATA_DIR / f'test_sets/pwkp/pwkp.valid.simp'],31    ('hsplit_test', 'orig'): DATA_DIR / f'test_sets/hsplit/hsplit.tok.src',32    ('hsplit_test', 'refs'): [DATA_DIR / f'test_sets/hsplit/hsplit.tok.{i+1}' for i in range(4)],33    ('wikisplit_test', 'orig'): DATA_DIR / f'test_sets/wikisplit/wikisplit.test.untok.orig',34    ('wikisplit_test', 'refs'): [DATA_DIR / f'test_sets/wikisplit/wikisplit.test.untok.split'],35    ('wikisplit_valid', 'orig'): DATA_DIR / f'test_sets/wikisplit/wikisplit.valid.untok.orig',36    ('wikisplit_valid', 'refs'): [DATA_DIR / f'test_sets/wikisplit/wikisplit.valid.untok.split'],37    ('googlecomp_test', 'orig'): DATA_DIR / f'test_sets/googlecomp/googlecomp.test.orig',38    ('googlecomp_test', 'refs'): [DATA_DIR / f'test_sets/googlecomp/googlecomp.test.comp'],39    ('googlecomp_valid', 'orig'): DATA_DIR / f'test_sets/googlecomp/googlecomp.valid.orig',40    ('googlecomp_valid', 'refs'): [DATA_DIR / f'test_sets/googlecomp/googlecomp.valid.comp'],41    ('qats_test', 'orig'): DATA_DIR / f'test_sets/qats/qats.test.orig',42    ('qats_test', 'refs'): [DATA_DIR / f'test_sets/qats/qats.test.simp'],43}44SYSTEM_OUTPUTS_DIR = DATA_DIR / "system_outputs"45SYSTEM_OUTPUTS_DIRS_MAP = {46    "turkcorpus_test": SYSTEM_OUTPUTS_DIR / "turkcorpus/test",47    "turkcorpus_valid": SYSTEM_OUTPUTS_DIR / "turkcorpus/valid",48    "pwkp_test": SYSTEM_OUTPUTS_DIR / "pwkp/test",49}50# Constants51VALID_TEST_SETS = list(set([test_set for test_set, language in TEST_SETS_PATHS.keys()])) + ['custom']52VALID_METRICS = [53    'bleu',54    'sari',55    'samsa',56    'fkgl',57    'sent_bleu',58    'f1_token',59    'sari_legacy',60    'sari_by_operation',61    'bertscore',62]...

controversy.py

Source:controversy.py

1from cpath import data_path2from data_generator.tokenizer_b import FullTokenizerWarpper3from evaluation import *4from models.cnn_predictor import CNNPredictor5from models.controversy import *6def eval_all_contrv():7    ams_X, ams_Y = amsterdam.get_dev_data(False)8    clue_X, clue_Y = controversy.load_clueweb_testset()9    guardian_X, guardian_Y = controversy.load_guardian()10    models = []11    #models.append(("CNN/Wiki", CNNPredictor("WikiContrvCNN")))12    models.append(("CNN/Wiki", CNNPredictor("WikiContrvCNN_sigmoid", "WikiContrvCNN")))13    #models.append(("tlm/wiki", get_wiki_doc_lm()))14    #models.append(("Bert/Wiki", BertPredictor("WikiContrv2009")))15    #models.append(("Bert/Wiki", BertPredictor("WikiContrv2009_only_wiki")))16    #models.append(("tlm/dbpedia", get_dbpedia_contrv_lm()))17    #models.append(("tlm/Guardian", get_guardian16_lm()))18    #models.append(("yw_may", get_yw_may()))19    #models.append(("Guardian2", get_guardian_selective_lm()))20    test_sets = []21    #test_sets.append(("Ams18", [ams_X, ams_Y]))22    test_sets.append(("Clueweb" ,[clue_X, clue_Y]))23    #test_sets.append(("Guardian", [guardian_X, guardian_Y]))24    for set_name, test_set in test_sets:25        dev_X, dev_Y = test_set26        print(set_name)27        for name, model in models:28            scores = model.score(dev_X)29            auc = compute_pr_auc(scores, dev_Y)30            #auc = compute_auc(scores, dev_Y)31            acc = compute_opt_acc(scores, dev_Y)32            prec = compute_opt_prec(scores, dev_Y)33            recall = compute_opt_recall(scores, dev_Y)34            f1 = compute_opt_f1(scores, dev_Y)35            print("{0}\t{1:.03f}\t{2:.03f}\t{3:.03f}\t{4:.03f}\t{5:.03f}".format(name, auc, prec, recall, f1, acc))36def dataset_stat():37    ams_X, ams_Y = amsterdam.get_dev_data(False)38    clue_X, clue_Y = controversy.load_clueweb_testset()39    guardian_X, guardian_Y = controversy.load_guardian()40    vocab_size = 3052241    vocab_filename = "bert_voca.txt"42    voca_path = os.path.join(data_path, vocab_filename)43    encoder = FullTokenizerWarpper(voca_path)44    test_sets = []45    test_sets.append(("Ams18", [ams_X, ams_Y]))46    test_sets.append(("Clueweb" ,[clue_X, clue_Y]))47    test_sets.append(("Guardian", [guardian_X, guardian_Y]))48    for set_name, test_set in test_sets:49        dev_X, dev_Y = test_set50        num_over_size = 051        length_list = []52        for doc in dev_X:53            tokens = encoder.encode(doc)54            if len(tokens) > 200:55                num_over_size += 156            length_list.append(len(tokens))57        print("{0} {1:.03f} {2:.03f}".format(set_name, num_over_size / len(dev_X), average(length_list)))58if __name__ == '__main__':...

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.