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

How to use save_file_as method in SeleniumBase

Best Python code snippet using SeleniumBase

util.py

Source:util.py

1from collections import defaultdict2from collections import OrderedDict3from os import path4import logging5import numpy as np6import torch7from properties import *8def init_logger(name='logger'):9    """Initialize and return a logger."""10    logger = logging.getLogger(name)11    logger.setLevel(logging.DEBUG)12    log_fmt = '%(asctime)s/%(name)s[%(levelname)s]: %(message)s'13    logging.basicConfig(format=log_fmt)14    return logger15def to_tensor(numpy_array):16    tensor = torch.from_numpy(numpy_array).float()17    if torch.cuda.is_available():18        tensor = tensor.cuda()19    return tensor20def to_variable(tensor, volatile=False):21    if torch.cuda.is_available():22        tensor = tensor.cuda()23    return torch.autograd.Variable(tensor, volatile)24def get_word_vectors_dicts(filename, dirname=DATA_DIR, save=False,25                           save_file_as='en_dict'):26    word2vec = OrderedDict()27    count = 028    print('Read ' + path.join(dirname, filename))29    with open(path.join(dirname, filename), 'r', encoding='utf-8') as f:30        N, D = f.readline().strip().split()31        D = int(D)32        for count, line in enumerate(f.readlines()):33            row = line.strip().split(' ')34            word = row[0]35            vec = np.array(row[1:]).astype(np.float)36            if len(vec) == D:37                word2vec[word] = vec38            count += 139            if count == top_frequent_words:40                break41    if save:42        np.save(path.join(dirname, save_file_as + '.npy'), word2vec)43    return word2vec44def get_validation_set(filename, dirname=DATA_DIR, save=False, save_file_as='validation'):45    true_dict = OrderedDict()46    print('Read ' + path.join(dirname, filename))47    with open(path.join(dirname, filename), 'r', encoding='utf-8') as f:48        rows = f.readlines()49        for row in rows:50            split_row = row.split(" ")51            key = split_row[0]52            value = split_row[1].rstrip("\n")53            if key not in true_dict.keys():54                true_dict[key] = []55                true_dict[split_row[0]].append(value)56    if save:57        np.save(path.join(dirname, save_file_as + '.npy'), true_dict)58    return true_dict59def get_frequencies():60    src = np.load(path.join(DATA_DIR, 'src.freq.npy'))61    trg = np.load(path.join(DATA_DIR, 'tgt.freq.npy'))62    return src, trg63class WeightedSampler():64    def __init__(self, weights, replacement=True):65        """Initialize a sampler.66        Arguments:67        - weights (list)   : a list of weights, not necessary summing up to one68        - replacement (bool): if ``True``, samples are drawn with replacement.69        """70        self.weights = torch.DoubleTensor(weights)71        self.replacement = replacement72    def get_iterator(self, batch_size):73        """Generate a batch of samples infinitely."""74        while True:75            yield torch.multinomial(self.weights, batch_size, self.replacement)76        77def downsample_frequent_words(counts, thresh=1e-3):78    """Discount frequent words."""79    total_count = counts.sum()80    indices_zero = counts == 081    counts[indices_zero] = 1.0  # avoid a numerical error82    threshold_count = float(thresh * total_count)83    new_weights = (np.sqrt(counts / threshold_count) + 1) * (threshold_count / counts)84    new_weights = np.maximum(new_weights, 1.0)85    new_weights *= counts86    new_weights[indices_zero] = 0.087    return new_weights / new_weights.sum()88class Utils:89    def __init__(self, params):90        self.data_dir = params.data_dir91        self.src_file = params.src_file92        self.tgt_file = params.tgt_file93        self.src_freq_file = params.src_freq_file94        self.tgt_freq_file = params.tgt_freq_file95        self.validation_file = params.validation_file96        self.top_frequent_words = params.top_frequent_words97    def run(self):98        print("Reading source word embeddings...")99        word2vec_src, _ = self.save_word_vectors(self.src_file, self.src_freq_file,100                                                 save=True, save_file_as='src')101        print("Done.")102        print(word2vec_src.shape)103        print("Reading target word embeddings...")104        word2vec_tgt, _ = self.save_word_vectors(self.tgt_file, self.tgt_freq_file,105                                                 save=True, save_file_as='tgt')106        print("Done.")107        print(word2vec_tgt.shape)108        print("Reading validation file...")109        self.save_validation_set(self.validation_file, save=True)110        print("Done.")111        print("Constructing source word-id map...")112        self.save_word_ids_dicts(self.src_file, save=True, save_file_as='src_ids')113        print("Done.")114        print("Constructing target word-id map...")115        self.save_word_ids_dicts(self.tgt_file, save=True, save_file_as='tgt_ids')116        print("Everything Done.")117    def save_word_vectors(self, emb_file, freq_file=None, save=False, save_file_as='src'):118        word2freq = defaultdict(int)119        if freq_file:120            with open(path.join(self.data_dir, freq_file), 'r', encoding='utf-8') as f:121                for line in f:122                    word, freq = line.strip().split(' ')123                    word2freq[word] = int(freq)124        embeddings, freqs = [], []125        print('Read ' + path.join(self.data_dir, emb_file))126        with open(path.join(self.data_dir, emb_file), 'r', encoding='utf-8') as f:127            N, D = f.readline().strip().split()128            D = int(D)129            for count, line in enumerate(f, start=1):130                row = line.strip().split(' ')131                word = row[0]132                vec = np.array(row[1:]).astype(np.float)133                if len(vec) == D:134                    embeddings.append(vec)135                    freqs.append(word2freq[word])136                if count == top_frequent_words:137                    break138            if save:139                np.save(path.join(self.data_dir, save_file_as + '.npy'),140                        np.array(embeddings))141                np.save(path.join(self.data_dir, save_file_as + '.freq.npy'),142                        np.array(freqs))143        return np.array(embeddings), freqs144    def save_word_ids_dicts(self, file, save=False, save_file_as='src_ids'):145        word2id = {}146        count = 0147        with open(path.join(self.data_dir, file), 'r', encoding='utf-8') as f:148            _, dim = f.readline().strip().split()149            dim = int(dim)150            for row in f:151                split_row = row.strip().split(" ")152                vec = np.array(split_row[1:]).astype(np.float)153                if len(vec) == dim:154                    word2id[split_row[0]] = count155                count += 1156                if count == self.top_frequent_words:157                    break158        if save:159            np.save(self.data_dir + save_file_as + '.npy', word2id)160        return word2id161    def save_validation_set(self, file, save=False, save_file_as='validation'):162        true_dict = {}163        with open(self.data_dir + file, 'r', encoding='utf-8') as f:164            rows = f.readlines()165            for row in rows:166                split_row = row.split(" ")167                key = split_row[0]168                value = split_row[1].rstrip("\n")169                if key not in true_dict.keys():170                    true_dict[key] = []171                true_dict[split_row[0]].append(value)172        if save:173            np.save(self.data_dir + save_file_as + '.npy', true_dict)174        return true_dict175def load_npy_one(data_dir, fname):176    return np.load(data_dir + fname).item()177def load_npy_two(data_dir, src_fname, tgt_fname, dict=False):178    if dict:179        x = np.load(data_dir + src_fname).item()180        y = np.load(data_dir + tgt_fname).item()181    else:182        x = np.load(data_dir + src_fname)183        y = np.load(data_dir + tgt_fname)184    return x, y185# Validation set in a dictionary form {src_wrd: [tgt_wrd_1, tgt_wrd_2, ...]}186def get_validation_set_ids(data_dir, validation_fname='validation.npy'):187    val_dict = load_npy_one(data_dir, validation_fname)188    src_ids, tgt_ids = load_npy_two(data_dir, 'src_ids.npy', 'tgt_ids.npy', dict=True)189    val_dict_ids = {}190    for src_wrd, tgt_list in val_dict.items():191        val_dict_ids[src_ids[src_wrd]] = [tgt_ids[tgt_wrd] for tgt_wrd in tgt_list192                                          if tgt_wrd in tgt_ids]193    return val_dict_ids194def convert_to_embeddings(emb_array):195    emb_tensor = to_tensor(emb_array)196    v, d = emb_tensor.size()197    emb = torch.nn.Embedding(v, d)198    if torch.cuda.is_available():199        emb = emb.cuda()200    emb.weight.data.copy_(emb_tensor)201    emb.weight.requires_grad = False...

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.