def evaluate_multi(url, url2, time_lags=24):
cr = Crawling()
preds = utils.load_file(url)
preds = np.array(preds)
lt = len(preds)
labels = utils.load_file(url2)
labels = np.array(labels)
loss_mae0, loss_mae1 = 0.0, 0.0
loss_rmse0, loss_rmse1 = 0.0, 0.0
r2_0, r2_1 = 0.0, 0.0
for i, d in enumerate(preds):
lb_i = i * pr.strides + time_lags + 1
mae0, mse0, r2 = get_evaluation(d[:time_lags, :],
labels[lb_i:(lb_i + time_lags), :, 0])
# mae1, mse1 = get_evaluation(d[:time_lags,:,1], labels[lb_i:(lb_i+time_lags),:,1])
loss_rmse0 += mse0
# loss_rmse1 += mse1
loss_mae0 += mae0
# loss_mae1 += mae1
r2_0 += r2
loss_mae0 = loss_mae0 / lt * 300
loss_mae1 = loss_mae1 / lt * 300
loss_rmse0 = sqrt(loss_rmse0 / lt) * 300
loss_rmse1 = sqrt(loss_rmse1 / lt) * 300
r2_0 = r2_0 / lt
print("MAE: %.6f %.6f" % (loss_mae0, cr.ConcPM25(loss_mae0)))
print("RMSE: %.6f %.6f" % (loss_rmse0, cr.ConcPM25(loss_rmse0)))
print("R2 Score: %.6f" % r2_0)
def execute_gan(path, attention_url, url_weight, model, session, saver, batch_size, encoder_length, decoder_length, is_test, train_writer=None, offset=0):
#if restore and not is_test:
# tf.reset_default_graph()
# print(tf.get_default_graph())
#with tf.device('/%s' % p.device):
# model.init_ops(not is_test)
# #model.add_placeholders()
#trainable_vars = tf.trainable_variables()
#saver = tf.train.Saver(trainable_vars)
print("==> Loading dataset")
dataset = utils.load_file(path)
if dataset:
dataset = np.asarray(dataset, dtype=np.float32)
lt = len(dataset)
train, _ = utils.process_data_grid(lt, batch_size, encoder_length, decoder_length, True)
attention_data = None
if attention_url:
attention_data = utils.load_file(attention_url)
model.set_data(dataset, train, None, attention_data)
#with tf.Session(config=gpu_configs) as session:
#init = tf.global_variables_initializer()
#session.run(init)
model.assign_datasets(session)
if not is_test:
print("start training")
for epoch in xrange(100):
_ = model.run_epoch(session, train, offset + epoch, train_writer, train=True, verbose=False, stride=2)
saver.save(session, 'weights/%s.weights' % url_weight)
else:
# saver.restore(session, url_weight)
print('==> running model')
_, preds = model.run_epoch(session, train, train=False, verbose=False, shuffle=False, stride=2)
save_gan_preds(url_weight, preds)
def get_grammatical_data(train_filename,
test_filename,
dict_filename,
translate_emojis=True,
replace_slang=True,
lowercase=True):
# Load the train and test sets
print("Loading data...")
train_tokens = utils.load_file(path + "/res/tokens/tokens_" +
train_filename)
train_pos = utils.load_file(path + "/res/pos/pos_" + train_filename)
test_tokens = utils.load_file(path + "/res/tokens/tokens_" + test_filename)
test_pos = utils.load_file(path + "/res/pos/pos_" + test_filename)
if translate_emojis and replace_slang and lowercase:
save_path = path + "/res/data/finest_grammatical_"
else:
save_path = path + "/res/data/grammatical_"
# Clean the data and brind it to the most *grammatical* form possible
gramm_train = grammatical_clean(train_tokens,
train_pos,
path + "/res/" + dict_filename,
save_path + train_filename,
translate_emojis=translate_emojis,
replace_slang=replace_slang,
lowercase=lowercase)
gramm_test = grammatical_clean(test_tokens,
test_pos,
path + "/res/" + dict_filename,
save_path + test_filename,
translate_emojis=translate_emojis,
replace_slang=replace_slang,
lowercase=lowercase)
return gramm_train, gramm_test
def get_traffic(**kwargs):
global TRAFFIC_WRAPPER
# t0 = time.time()
if TRAFFIC_WRAPPER is None:
wrapperFile = 'wrappers/traffic_wrapper.json'
synonyms = load_synonyms('./datasets/sinonimos.csv')
words = load_words()
if os.path.isfile(wrapperFile):
with open(wrapperFile,'r+') as rwjson:
TRAFFIC_WRAPPER = ClassifierWrapper()
TRAFFIC_WRAPPER.jsonLoads(rwjson.read())
TRAFFIC_WRAPPER.dataset.dataset = list(load_file('./datasets/traffic2.csv'))
TRAFFIC_WRAPPER.synonyms = copy.deepcopy(synonyms)
TRAFFIC_WRAPPER.words = copy.deepcopy(words)
TRAFFIC_WRAPPER.dataset.synonyms = copy.deepcopy(synonyms)
TRAFFIC_WRAPPER.dataset.words = copy.deepcopy(words)
return TRAFFIC_WRAPPER
clf = kwargs.pop('clf', LogisticRegression(C=8.5))
dataWrapperDataset = list(load_file('./datasets/traffic2.csv'))
dataWrapper = DataWrapper(dataset=dataWrapperDataset,synonyms=copy.deepcopy(synonyms),words=copy.deepcopy(words))
dataWrapper.resolveMatrix()
wrapper = ClassifierWrapper(clf=clf,dataset=dataWrapper,synonyms=copy.deepcopy(synonyms),words=copy.deepcopy(words))
cross_validate = kwargs.pop('cross_validate', True)
if cross_validate:
wrapper.cross_validate()
wrapper.train()
# print time.time() - t0, "seconds from the multiclass classifier"
TRAFFIC_WRAPPER = wrapper
with open(wrapperFile, 'w') as rw_json:
json.dump(TRAFFIC_WRAPPER.toDict(), rw_json)
return TRAFFIC_WRAPPER
def main(data_path):
if path.exists(data_path + "/dict_char_en.pkl"):
dict_char_en = utils.load_file(data_path + "/dict_char_en.pkl")
else:
dict_char_en = generateCharacterDict(properties.en_char)
utils.save_file(data_path + "/dict_char_en.pkl", dict_char_en)
if path.exists(data_path + "/dict_char_vi.pkl"):
dict_char_vi = utils.load_file(data_path + "/dict_char_vi.pkl")
else:
dict_char_vi = generateCharacterDict(properties.vi_char)
utils.save_file(data_path + "/dict_char_vi.pkl", dict_char_vi)
if path.exists(data_path + "/dict_en.pkl"):
dict_en = utils.load_file(data_path + "/dict_en.pkl", True)
else:
dict_en = build_dictionary(data_path, properties.vocab_en)
utils.save_file(data_path + "/dict_en.pkl", dict_en, True)
if path.exists(data_path + "/dict_vi.pkl"):
dict_vi = utils.load_file(data_path + "/dict_vi.pkl", True)
else:
dict_vi = build_dictionary(data_path, properties.vocab_vi)
utils.save_file(data_path + "/dict_vi.pkl", dict_vi, True)
dataset_en, unknown_en = map_sentence_idx(
data_path + "/" + properties.train_en, dict_en, dict_char_en)
dataset_vi, unknown_vi = map_sentence_idx(
data_path + "/" + properties.train_vi, dict_vi, dict_char_vi)
utils.save_file(data_path + "/dataset_en.pkl", (dataset_en, unknown_en))
utils.save_file(data_path + "/dataset_vi.pkl", (dataset_vi, unknown_vi))
def read_data(args):
data_sources = []
header = ''
if (args.f != None):
if not isinstance(args.f, basestring):
parts = []
for afile in args.f:
part_of_data = utils.load_file(afile)
if args.e != None and args.e == 'y':
if header == '':
header = part_of_data[0]
part_of_data = part_of_data[1:len(part_of_data)]
parts.append(part_of_data.tolist())
if args.s != None and args.s == 'y':
parts.append("\n")
parts = [item for sublist in parts for item in sublist]
data_sources = array(parts)
else:
data_sources = utils.load_file(args.f)
output = []
if header != '':
output.append(header)
for item in data_sources:
output.append(item)
return output
def evaluate_transportation(url, url2, pred_length=8):
preds = utils.load_file(url)
preds = np.array(preds)
lt = len(preds)
labels = utils.load_file(url2)
labels = np.array(labels)
labels = labels.reshape(len(labels), 32, 32)
shape = np.shape(preds)
if preds.shape[-1] < pred_length:
print("data shape is ", preds.shape)
pred_length = preds[-1]
loss_mae0 = [0.0] * pred_length
loss_rmse0 = [0.0] * pred_length
r2_total = 0.0
for i, d in enumerate(preds):
# 8 is encoder_length
lb_i = i + 8
# labels[lb_i:(pred_length+lb_i),:,:]
for x in xrange(pred_length):
mae0, mse0, _ = get_evaluation(d[x, :, :], labels[lb_i + x, :, :])
# mae0, mse0, r2 = get_evaluation(d[0,:,:], labels[lb_i,:,:])
loss_rmse0[x] += mse0
loss_mae0[x] += mae0
# r2_total += r2
loss_mae0 = [(x / lt * 131) for x in loss_mae0]
loss_rmse0 = [(sqrt(x / lt) * 131) for x in loss_rmse0]
# r2_total = r2_total / lt
# print("MAE: %.6f" % loss_mae0)
# print("RMSE: %.6f" % loss_rmse0)
# print("R2 Score: %.6f" % r2_total)
print_accumulate_error(loss_mae0, loss_rmse0, pred_length, 0)
def get_data(vocabs=""):
print("==> Load Word Embedding")
word_embedding = utils.load_glove(use_index=True)
validation_data = []
training_data = []
if not vocabs:
non_words = utils.load_file(p.non_word, False)
for w in non_words:
w_ = w.replace('\n', '').split(' ')
validation_data.append(int(w_[-1]))
training_data = utils.sub(range(len(word_embedding)), validation_data)
else:
vocabs_set = utils.load_file(vocabs)
print("vc", len(vocabs_set))
training_data = [w for _, w in vocabs_set.iteritems()]
tm = range(len(word_embedding))
validation_data = list(utils.sub(set(tm), set(training_data)))
length = int(math.ceil(len(training_data) * 1.0 / p.compression_batch_size)) * p.compression_batch_size - len(training_data)
print('before', 'vd', len(validation_data), 'td', len(training_data))
if length:
add_on = np.random.choice(validation_data, length)
training_data += add_on.tolist()
validation_data = utils.sub(set(validation_data), set(add_on))
print('vd', len(validation_data), 'td', len(training_data))
# utils.save_file(p.glove_path, training_data)
return word_embedding, training_data, validation_data
def execute_gan(path, attention_url, label_path, url_weight, model, session, saver, batch_size, encoder_length, decoder_length, is_test, train_writer=None, offset=0, gpu_nums=1):
print("==> Loading dataset")
dataset = utils.load_file(path)
if dataset:
dataset = np.asarray(dataset, dtype=np.float32)
lt = len(dataset)
train, _ = utils.process_data_grid(lt, batch_size, encoder_length, decoder_length, True)
attention_data = None
if attention_url:
attention_data = utils.load_file(attention_url)
labels = None
if label_path:
labels = utils.load_file(label_path)
model.set_data(dataset, train, None, attention_data, labels)
model.assign_datasets(session)
if not is_test:
print('==> starting training')
suffix = p.weight_saving_break
for epoch in xrange(p.total_iteration):
_ = model.run_epoch(session, train, offset + epoch, train_writer, train=True, verbose=False, stride=4)
tmp_e = epoch + 1
if tmp_e % 10 == 0:
suffix = math.ceil(float(tmp_e) / p.weight_saving_break)
# utils.update_progress((epoch + 1) * 1.0 / p.total_iteration)
saver.save(session, 'weights/%s_%i.weights' % (url_weight, suffix))
saver.save(session, 'weights/%s_%i.weights' % (url_weight, suffix))
else:
# saver.restore(session, url_weight)
print('==> running model')
_, preds = model.run_epoch(session, train, train=False, verbose=False, shuffle=False, stride=2)
save_gan_preds(url_weight, preds)
def convert_vocab_to_text(vocabs):
vocab_str = ""
length = len(vocabs)
i = 0
vocab_idx = dict()
vocab_lst = list()
idx_file = '%s/%s' % (folder, 'vocabs_idx.pkl')
if u.check_file(idx_file):
vocab_idx = u.load_file(idx_file)
else:
for key, value in vocabs.iteritems():
vocab_idx[value] = key
u.save_file(idx_file, vocab_idx)
lst_file = '%s/%s' % (folder, 'vocabs_list.pkl')
if u.check_file(lst_file):
vocab_lst = u.load_file(lst_file)
else:
for key in sorted(vocab_idx.iterkeys()):
vocab_lst.append(vocab_idx[key])
u.save_file(lst_file, vocab_lst)
regex = RegexpTokenizer(r'\w+')
for w in vocab_lst:
words = regex.tokenize(w)
if len(words) != 0:
w_ = '_'.join(words)
i += 1
if i % 10000 == 0:
print('Processed %i' % i)
# break
if i == length:
vocab_str += '%s' % w_
else:
vocab_str += '%s\n' % w_
return vocab_str
def read_data(args):
data_sources = []
header = ''
if (args.f != None):
if not isinstance(args.f, basestring):
parts = []
for afile in args.f:
part_of_data = utils.load_file(afile)
if args.e != None and args.e == 'y':
if header == '':
header = part_of_data[0]
part_of_data = part_of_data[1:len(part_of_data)]
parts.append(part_of_data.tolist())
if args.s != None and args.s == 'y':
parts.append("\n")
parts = [item for sublist in parts for item in sublist]
data_sources = array(parts)
else:
data_sources = utils.load_file(args.f)
output = []
if header != '':
output.append(header)
for item in data_sources:
output.append(item)
return output
def build_emoji_sentiment_dictionary():
new_emoji_sentiment_filename = path + "/res/emoji/emoji_sentiment_dictionary.txt"
if not os.path.exists(new_emoji_sentiment_filename):
filename = path + "/res/emoji/emoji_sentiment_raw.txt"
emojis = utils.load_file(filename)[1:]
lines = []
for line in emojis:
line = line.split(",")
emoji = line[0]
occurences = line[2]
negative = float(line[4]) / float(occurences)
neutral = float(line[5]) / float(occurences)
positive = float(line[6]) / float(occurences)
description = line[7]
lines.append(
str(emoji) + "\t" + str(negative) + "\t" + str(neutral) +
"\t" + str(positive) + "\t" + description.lower())
utils.save_file(lines, new_emoji_sentiment_filename)
emoji_sentiment_data = utils.load_file(new_emoji_sentiment_filename)
emoji_sentiment_dict = {}
for line in emoji_sentiment_data:
line = line.split("\t")
# Get emoji characteristics as a list [negative, neutral, positive, description]
emoji_sentiment_dict[line[0]] = [line[1], line[2], line[3], line[4]]
return emoji_sentiment_dict
def exe(word_vectors_file, vector_preloaded_path, train_path, dev_path,
test_path, hsi, hso, maxlen, pep, fep, ppat, fpat, plr, flr, mix):
global word_vectors, vocabs
if os.path.exists(train_path) and os.path.exists(
dev_path) and os.path.exists(test_path):
train = utils.load_file(train_path)
dev = utils.load_file(dev_path)
test = utils.load_file(test_path)
else:
raise NotImplementedError()
if word_vectors is None or vocabs is None:
word_vectors, vocabs = utils.loadWordVectors(word_vectors_file,
vector_preloaded_path)
if not maxlen:
maxlen = properties.maxlen
lstm = Model(word_vectors,
hidden_sizes=[hsi, hso],
epochs=pep,
patience=ppat,
learning_rate=plr)
lstm_params = lstm.train(train, dev, test, maxlen)
if mix is 'Y':
combined = LSTM_CNN(word_vectors,
hidden_sizes=[hsi, hso],
epochs=fep,
lstm_params=lstm_params)
combined.train(train, dev, test, maxlen)
def evaluate_single_pred(url, url2, decoder_length=8):
cr = Crawling()
data = utils.load_file(url)
if type(data) is list:
data = np.asarray(data)
lt = data.shape[0] * data.shape[1]
data = np.reshape(data, (lt, 25))
dtl = len(data)
labels = utils.load_file(url2)
labels = np.asarray(labels)
loss_mae = 0.0
loss_rmse = 0.0
r2_total = 0.0
for i, d in enumerate(data):
pred_t = np.asarray(d).flatten()
lb_i = i * pr.strides + 24
lbt = labels[lb_i:(lb_i + decoder_length), :, 0]
lbg = lbt[decoder_length - 1, :].flatten()
mae, mse, r2 = get_evaluation(pred_t, lbg)
loss_mae += mae
loss_rmse += mse
r2_total += r2
utils.update_progress((i + 1.0) / dtl)
loss_mae = loss_mae / lt * 300
loss_rmse = sqrt(loss_rmse / lt) * 300
r2_total = r2_total / lt
print("MAE: %.6f %.6f" % (loss_mae, cr.ConcPM25(loss_mae)))
print("RMSE: %.6f %.6f" % (loss_rmse, cr.ConcPM25(loss_rmse)))
print("R2 score: %.6f" % r2_total)
def prepare_data(shuffle=False, labels_to_categorical=True):
path = os.getcwd()[:os.getcwd().rfind("/")]
to_write_filename = path + "/stats/data_prep_for_lstm_visualization.txt"
utils.initialize_writer(to_write_filename)
train_filename = "train.txt"
test_filename = "test.txt"
tokens_filename = "clean_original_" # other types of tokens to experiment with in /res/tokens/
data_path = path + "/res/tokens/tokens_"
# Load the data
train_data = utils.load_file(data_path + tokens_filename + train_filename)
test_data = utils.load_file(data_path + tokens_filename + test_filename)
if shuffle:
train_data = utils.shuffle_words(train_data)
test_data = utils.shuffle_words(test_data)
print("DATA IS SHUFFLED")
# Load the labels
train_labels = [
int(l) for l in utils.load_file(path + "/res/datasets/ghosh/labels_" +
train_filename)
]
test_labels = [
int(l) for l in utils.load_file(path + "/res/datasets/ghosh/labels_" +
test_filename)
]
# Get the max length of the train tweets
max_tweet_length = utils.get_max_len_info(train_data)
# Convert all tweets into sequences of word indices
tokenizer, train_indices, test_indices = utils.encode_text_as_word_indexes(
train_data, test_data, lower=True)
vocab_size = len(tokenizer.word_counts) + 1
word_to_index = tokenizer.word_index
print("There are %s unique tokens." % len(word_to_index))
# Pad sequences with 0s (can do it post or pre - post works better here)
x_train = pad_sequences(train_indices,
maxlen=max_tweet_length,
padding="post",
truncating="post",
value=0.)
x_test = pad_sequences(test_indices,
maxlen=max_tweet_length,
padding="post",
truncating="post",
value=0.)
# Transform the output into categorical data or just keep it as it is (in a numpy array)
if labels_to_categorical:
train_labels = to_categorical(np.asarray(train_labels))
test_labels = to_categorical(np.asarray(test_labels))
else:
train_labels = np.array(train_labels)
test_labels = np.array(test_labels)
return x_train, train_labels, x_test, test_labels, vocab_size, tokenizer, max_tweet_length
def evaluate_sp(url, url2, decoder_length=24, is_grid=True, grid_eval=True):
cr = Crawling()
map_ = heatmap.build_map()
data = utils.load_file(url)
if type(data) is list:
data = np.asarray(data)
if len(data.shape) == 4:
lt = data.shape[0] * data.shape[1]
else:
lt = data.shape[0]
if is_grid:
data = np.reshape(data, (lt, data.shape[-2], 25, 25))
else:
data = np.reshape(data, (lt, data.shape[-2], 25))
labels = utils.load_file(url2)
labels = np.asarray(labels)
loss_mae = 0.0
loss_rmse = 0.0
r2_total = 0.0
for i, d in enumerate(data):
d = d[:decoder_length, :, :]
pred_t = []
if is_grid:
for d_ in d:
d_t = heatmap.clear_interpolate_bound(np.asarray(d_), map_)
pred_t.append(d_t)
else:
if grid_eval:
for d_ in d:
d_t = heatmap.fill_map(d_, map_)
pred_t.append(d_t)
else:
pred_t = d
lb_i = i * pr.strides + 24
lbt = labels[lb_i:(lb_i + decoder_length), :, 0]
if grid_eval:
lbg = []
for x in lbt:
x_l = heatmap.fill_map(x, map_)
lbg.append(x_l)
lbg = np.asarray(lbg)
lbg = lbg.flatten()
else:
lbg = lbt.flatten()
pred_t = np.asarray(pred_t)
pred_t = pred_t.flatten()
mae, mse, r2 = get_evaluation(pred_t, lbg)
loss_mae += mae
loss_rmse += mse
r2_total += r2
utils.update_progress((i + 1.0) / lt)
loss_mae = loss_mae / lt * 300
loss_rmse = sqrt(loss_rmse / lt) * 300
r2_total = r2_total / lt
print("MAE: %.6f %.6f" % (loss_mae, cr.ConcPM25(loss_mae)))
print("RMSE: %.6f %.6f" % (loss_rmse, cr.ConcPM25(loss_rmse)))
print("R2 Score: %.6f" % r2_total)
def accuracy_metric(sample_list_file, result_file):
sample_list = load_file(sample_list_file)
group = {
'CW': [],
'CH': [],
'TN': [],
'TC': [],
'DC': [],
'DL': [],
'DO': []
}
for id, row in sample_list.iterrows():
qns_id = str(row['video']) + '_' + str(row['qid'])
qtype = str(row['type'])
#(combine temporal qns of previous and next as 'TN')
if qtype == 'TP': qtype = 'TN'
group[qtype].append(qns_id)
preds = load_file(result_file)
group_acc = {'CW': 0, 'CH': 0, 'TN': 0, 'TC': 0, 'DC': 0, 'DL': 0, 'DO': 0}
group_cnt = {'CW': 0, 'CH': 0, 'TN': 0, 'TC': 0, 'DC': 0, 'DL': 0, 'DO': 0}
overall_acc = {'C': 0, 'T': 0, 'D': 0}
overall_cnt = {'C': 0, 'T': 0, 'D': 0}
all_acc = 0
all_cnt = 0
for qtype, qns_ids in group.items():
cnt = 0
acc = 0
for qid in qns_ids:
cnt += 1
answer = preds[qid]['answer']
pred = preds[qid]['prediction']
if answer == pred:
acc += 1
group_cnt[qtype] = cnt
group_acc[qtype] += acc
overall_acc[qtype[0]] += acc
overall_cnt[qtype[0]] += cnt
all_acc += acc
all_cnt += cnt
for qtype, value in overall_acc.items():
group_acc[qtype] = value
group_cnt[qtype] = overall_cnt[qtype]
for qtype in group_acc:
print(map_name[qtype], end='\t')
print('')
for qtype, acc in group_acc.items():
print('{:.2f}'.format(acc * 100.0 / group_cnt[qtype]), end='\t')
print('')
print('Acc: {:.2f}'.format(all_acc * 100.0 / all_cnt))
def main():
parser = get_parser()
args = parser.parse_args()
if args.doc:
print __doc__
sys.exit()
g = geosearchclass.GeoSearchClass()
if args.params:
print 'Using parameters from ' + str(args.params)
# turn parameter file into dictionary
g.set_params_from_file(args.params)
if args.address:
print "Finding geocoordates for address:\n{}".format(args.address)
coords = geo_converter.get_geocoords_from_address(args.address)
if coords:
g.latitude = coords[0]
print "Found this latitude:"
print g.latitude
g.longitude = coords[1]
print "Found this longitude:"
print g.longitude
else:
print "Failed to find coordinates. Exiting."
sys.exit()
if args.input:
text = utils.load_file(args.input)
tokens = utils.tokenize_normal_words(text)
for_poem = utils.filter_words(tokens)
else:
for_poem = get_default_words()
if args.markov:
if args.input:
raise StandardError("Can only input a single text file. \
use --markov <your_text_file.txt>")
else:
text = utils.load_file(args.markov)
# ngram = ngrams.make_ngram(text, 2)
ngram = ngrams.make_bigram_trigram_dictionary(text)
formatted_poem = create_poem(g, for_poem, ngram)
else:
formatted_poem = create_poem(g, for_poem)
if args.output:
print '\nwriting formatted poem to ' + str(args.output)
output_file = args.output
else:
print "\nwriting formatted poem to poem.txt"
output_file = "poem.txt"
utils.save_file(output_file, formatted_poem)
def load_trained_params(self):
lstm = utils.load_file('lstm_cb.txt')
hidden_lstm = utils.load_file('hidden_cb.txt')
hidden_relu_lstm = utils.load_file('hidden_relu_cb.txt')
full_connect_lstm = utils.load_file('full_connect_cb.txt')
convs = list()
for x in range(len(self.filter_sizes)):
conv = utils.load_file('convolution_%s.txt' % x)
convs.append(conv)
return lstm, hidden_lstm, hidden_relu_lstm, full_connect_lstm, convs
def get_filtered_clean_data(train_filename, test_filename):
# Loading the train and test sets
print("Loading data...")
train_tokens = utils.load_file(path + "/res/data/" + train_filename)
test_tokens = utils.load_file(path + "/res/data/" + test_filename)
filtered_train_tokens = ulterior_clean(
train_tokens, path + "/res/data/filtered_" + train_filename)
filtered_test_tokens = ulterior_clean(
test_tokens, path + "/res/data/filtered_" + test_filename)
return filtered_train_tokens, filtered_test_tokens
def evaluate_lstm(url,
url2,
decoder_length=24,
forecast_factor=0,
is_classify=False):
data = utils.load_file(url)
if type(data) is list:
data = np.asarray(data)
lt = data.shape[0] * data.shape[1]
data = np.reshape(data, (lt, data.shape[-1]))
if decoder_length > data.shape[-1]:
decoder_length = data.shape[-1]
dtl = len(data)
labels = utils.load_file(url2)
labels = np.asarray(labels)
if not is_classify:
loss_mae = [0.0] * decoder_length
loss_rmse = [0.0] * decoder_length
else:
acc = 0.
#: r2_total = 0.0
cr = Crawling()
for i, d in enumerate(data):
if decoder_length < data.shape[-1]:
pred_t = d[:decoder_length]
else:
pred_t = d
lb_i = i * pr.strides + 24
lbt = np.mean(labels[lb_i:(lb_i + decoder_length), :, forecast_factor],
axis=1)
a = 0.
for t_i, (p, l) in enumerate(zip(pred_t, lbt)):
if not is_classify:
# mae, mse, _ = get_evaluation(p, l)
mae = abs(cr.ConcPM10(p * 300) - cr.ConcPM10(l * 300))
loss_mae[t_i] += mae
# loss_rmse[t_i] += mse
else:
a += classify_data(pred_t, lbt, forecast_factor)
if is_classify:
a = a / decoder_length
acc += a
# r2_total += r2
utils.update_progress((i + 1.0) / dtl)
if not is_classify:
loss_mae = np.array(loss_mae) / lt
# loss_rmse = [sqrt(x / lt) * 300 for x in loss_rmse]
# print("R2 score: %.6f" % r2_total)
print_accumulate_error(loss_mae,
loss_rmse,
decoder_length,
forecast_factor=forecast_factor)
else:
acc = acc / lt * 100
print("accuracy %.4f" % acc)
def get_strict_data(train_filename, test_filename):
# Load the train and test sets
print("Loading data...")
train_tweets = utils.load_file(path + "/res/data/" + train_filename)
test_tweets = utils.load_file(path + "/res/data/" + test_filename)
# Initial clean of data
strict_tweets_train = strict_clean(
train_tweets, path + "/res/data/strict_" + train_filename)
strict_tweets_test = strict_clean(
test_tweets, path + "/res/data/strict_" + test_filename)
return strict_tweets_train, strict_tweets_test
def __init_docs(self, file_names):
docs = [] # type:List[Document]
for f_name in file_names:
# 读取text文档
text = load_file(self.root_path, f_name, "txt") # type:str
# 读取entities和ent_pairs
ann_data = load_file(self.root_path, f_name, "ann")
entities, entity_pairs = self.__get_entities_and_pairs(
ann_data) # type:NamedEntitySet,List[EntityPair]
d = Document(f_name, self.root_path, text, entities, entity_pairs)
docs.append(d)
self._docs = docs
def transform(self, X, **transform_params):
docs_topics_vectors = []
lda_model = load_file("models/LDAbow_fbpac.pickle")
lda_dictionary = load_file("models/LDAdict_fbpac.pickle")
for doc in X:
try:
bow_vector = lda_dictionary.doc2bow(pre_process(doc))
docs_topics_vectors.append(lda_model[bow_vector])
except Exception as e:
print(e)
print("Error in computing topic vector")
n, nx, ny = np.array(docs_topics_vectors).shape
d2_all_docs = np.array(docs_topics_vectors).reshape((n, nx * ny))
return d2_all_docs[:, 1::2]
def main(urls, file=False):
global loaded
a_load = utils.load_file('cached.pkl')
if a_load:
loaded = a_load
else:
loaded = dict()
if file:
urls = utils.load_file(urls)
# bad = utils.load_file('sitemap_bad.txt')
elif urls:
urls = urls.split(',')
scrape_list(urls)
utils.save_file('cached.pkl', loaded)
def get_clean_dl_data(train_filename, test_filename, word_list):
vocab_filename = "dnn_vocabulary_" + train_filename
# Load the train and test sets
print("Loading data...")
train_tweets = utils.load_file(path + "/res/tokens/tokens_" +
train_filename)
test_tweets = utils.load_file(path + "/res/tokens/tokens_" + test_filename)
vocabulary = build_vocabulary_for_dnn_tasks(
path + "/res/vocabulary/" + vocab_filename, train_tweets)
clean_train_tweets, train_indices = vocabulary_filtering(
vocabulary, train_tweets)
clean_test_tweets, test_indices = vocabulary_filtering(
vocabulary, test_tweets)
return clean_train_tweets, train_indices, clean_test_tweets, test_indices, len(
vocabulary)
def convert_data_to_grid(url, out_url, url_att="", out_url_att="", part=1):
grid = heatmap.build_map(pr.map_size)
data = utils.load_file(url)
lt = len(data)
attention_data = None
att_part = None
print(url_att)
if url_att:
attention_data = utils.load_file(url_att)
alt = len(attention_data)
if lt != alt:
raise ValueError(
"Attention & Main Data need same length while %s and %s" %
(lt, alt))
data = zip(data, attention_data)
att_part = []
res = []
if part != 1:
bound = int(math.ceil(float(lt) / part))
else:
bound = lt
for i, row in enumerate(data):
if url_att:
t, a = row
else:
t = row
if i and (i % bound) == 0:
p_i = i / bound
out_url_name = out_url + "_" + str(p_i)
utils.save_file(out_url_name, res)
if url_att:
att_out_url_name = out_url_att + "_" + str(p_i)
utils.save_file(att_out_url_name, att_part)
res = []
att_part = []
g = heatmap.fill_map(t, grid)
res.append(g)
if url_att:
att_part.append(a)
utils.update_progress(float(i) / lt)
if part == 1:
out_url_name = out_url
else:
out_url_name = out_url + "_" + str(part)
utils.save_file(out_url_name, res)
if url_att:
att_out_url_name = out_url_att + "_" + str(part)
utils.save_file(att_out_url_name, att_part)
def load_multi_data(args):
header = []
data = {}
if (args.f != None):
if not isinstance(args.f, basestring):
for afile in args.f:
file_lines = utils.load_file(afile)
count = 0
for line in file_lines:
org_line = line
line = line.rsplit(',')
if (len(line) <= 1):
line = org_line.rsplit("\t")
if (count == 0):
if (len(header) > 0):
header.append(', '.join(line[1:len(line)]))
else:
header.append(', '.join(line))
count = count + 1
continue
if (line[0] not in data):
data[line[0]] = []
data[line[0]].append(', '.join(line[1:len(line)]))
return (header, data)
def pdb_to_lh5(traj, field):
path = getattr(traj, field)
data = load_file(path)
new_fn = os.path.splitext(path)[0] + '.lh5'
save_file(new_fn, data)
os.unlink(path)
setattr(traj, field, new_fn)
def read_data(args):
data_sources = []
header = ""
if args.f != None:
parts = []
afile = args.f
part_of_data = utils.load_file(afile)
if args.e != None and args.e == "y":
if header == "":
header = part_of_data[0]
part_of_data = part_of_data[1 : len(part_of_data)]
part_of_data = calculat_hsv_model(part_of_data, args.t)
parts.append(part_of_data)
parts = [item for sublist in parts for item in sublist]
data_sources = array(parts)
output = []
if header != "":
output.append(header)
for item in data_sources:
output.append(item)
return output
def main(args):
# Load file content.
content = load_file(args.input, encoding=args.encoding)
# Clean content.
cleaned = clean(content, args.pattern)
# Save cleaned content.
save_file(args.output, cleaned, encoding=args.encoding)
def split_hashtag_long_version(hashtag):
word_file = path + "/res/word_list.txt"
word_list = utils.load_file(word_file).split()
word_dictionary = list(set(words.words()))
for alphabet in "bcdefghjklmnopqrstuvwxyz":
word_dictionary.remove(alphabet)
all_poss = split_hashtag_to_words_all_possibilities(
hashtag.lower(), word_dictionary)
max_p = 0
min_len = 1000
found = False
best_p = []
for poss in all_poss:
counter = 0
for p in poss:
if p in word_list:
counter += 1
if counter == len(poss) and min_len > counter:
found = True
min_len = counter
best_p = poss
else:
if counter > max_p and not found:
max_p = counter
best_p = poss
best_p_v2 = split_hashtag(hashtag, word_list)
if best_p != [] and best_p_v2 != []:
split_words = best_p if len(best_p) < len(best_p_v2) else best_p_v2
else:
if best_p == [] and best_p_v2 == []:
split_words = [hashtag]
else:
split_words = best_p if best_p_v2 == [] else best_p_v2
split_words = ['#' + str(s) for s in split_words]
return split_words
def read_data(args):
[operator, comparing_field_index, threshold] = args.t.rsplit(',')
data_sources = []
header = ''
if (args.f != None):
parts = []
afile = args.f
part_of_data = utils.load_file(afile)
if args.e != None and args.e == 'y':
if header == '':
header = part_of_data[0]
part_of_data = part_of_data[1:len(part_of_data)]
part_of_data = filter_content(part_of_data, operator, comparing_field_index, threshold)
parts.append(part_of_data)
parts = [item for sublist in parts for item in sublist]
data_sources = array(parts)
output = []
if header != '':
output.append(header)
for item in data_sources:
output.append(item)
return output
def read_data(args):
data_sources = []
header = ''
if (args.f != None):
parts = []
afile = args.f
part_of_data = utils.load_file(afile)
if args.e != None and args.e == 'y':
if header == '':
header = part_of_data[0]
part_of_data = part_of_data[1:len(part_of_data)]
part_of_data = calculat_facedetection_model(part_of_data, args.t, args.c)
parts.append(part_of_data)
parts = [item for sublist in parts for item in sublist]
data_sources = array(parts)
output = []
if header != '':
output.append(header)
for item in data_sources:
output.append(item)
return output
def on_return(self, task):
"""Called by main thread on the return of data from the workers.
Post-processing"""
logger.info('Retrieved task %s', task.tag)
traj = Session.query(models.Trajectory).get(int(task.tag))
try:
# save lh5 version of the trajectory
conf = load_file(self.project.pdb_topology_file)
coordinates = msmbuilder.Trajectory.load_trajectory_file(str(traj.dry_xtc_fn), Conf=conf)
save_file(traj.lh5_fn, coordinates)
except Exception as e:
logger.error('When postprocessing %s, convert to lh5 failed!', traj)
logger.exception(e)
raise
# convert last_wet_snapshot to lh5
pdb_to_lh5(traj, 'last_wet_snapshot_fn')
pdb_to_lh5(traj, 'init_pdb_fn')
traj.host = task.host
traj.returned_time = datetime.now()
traj.length = len(coordinates)
logger.info('Finished converting new traj to lh5 sucessfully')
def on_open1_activate(self, widget=None, file=''):
self.on_quit2_activate()# this takes care of saving content yes/no
if not file:
dlg = FileDialog(action='open',title=_("Open GvR world"),ext='wld')
response = dlg.run()
if response == Gtk.ResponseType.OK:
file = dlg.get_filename()
if os.path.splitext(file)[1] != '.wld':
self.show_error(_("Selected path is not a world file"))
dlg.destroy()
return
elif response == Gtk.ResponseType.CANCEL:
self.logger.debug('Closed, no files selected')
dlg.destroy()
return
dlg.destroy()
txt = utils.load_file(file)
if txt:
self.set_text(file,txt)
self.parent.on_button_reload()
return
def on_open1_activate(self, widget=None,file=''):
if not file:
dlg = FileDialog(action='open',title=_("Open GvR program"),ext='gvr')
response = dlg.run()
if response == Gtk.ResponseType.OK:
file = dlg.get_filename()
if os.path.splitext(file)[1] != '.gvr':
self.show_error(_("Selected path is not a program file"))
dlg.destroy()
return
elif response == Gtk.ResponseType.CANCEL:
self.logger.debug('Closed, no files selected')
dlg.destroy()
return
dlg.destroy()
txt = utils.load_file(file)
if txt:
self.set_text(file,txt)
for b in ('execute','step','abort'):
self.parent._set_sensitive_button(b,True)
return
def read_file(afile, fields, filters, data_sources, data_labels, padding = None):
part_of_data = utils.load_file(afile)
part_of_data = part_of_data[1:len(part_of_data)]
content = {}
count = 0
for line in part_of_data:
line_fields = line.rsplit(',')
if (len(line_fields) == 1):
line_fields = line.rsplit("\t")
selected_line = ''
if (len(filters) == 0 or data_labels[count] in filters):
selected_line = [line_fields[int(index)] for index in fields]
else:
if padding != None:
selected_line = padding
else:
selected_line = ['0' for index in fields]
#content.append(','.join(selected_line))
content[line_fields[0]] = ','.join(selected_line)
count += 1
output_content = []
for item in data_sources:
if (item in content):
output_content.append(content[item])
else:
print("key " + item + " is not in input file.")
return "\n".join(output_content)
def main():
parser = argparse.ArgumentParser(description = 'Exporting data matrix from HIT summary result.')
parser.add_argument('-f', action = 'append', help = 'The CSV files.')
parser.add_argument('-c', help = 'The exporting columns separated with comma.')
parser.add_argument('-o', help = 'The output file.')
parser.add_argument('-t', help = 'The types used to filter out data row.')
parser.add_argument('-p', default = '0', help = 'The padding for filtered rows.')
parser.add_argument('-d', help = 'The data source file.')
args = parser.parse_args()
data_sources = []
data_labels = []
data_ids = []
if (args.d != None):
data_sources = utils.load_file(args.d)
data_metainfo = regex_datasource(data_sources)
# data_labels: flickr high interesting 1, flickr low interesting 2, pinterest [3, 4, 5]
data_labels = data_metainfo[0]
# data_ids: (flickr, pinterest) image id
data_ids = data_metainfo[1]
output = read_data(args, data_sources, data_labels)
if (args.o != None):
utils.write_file(output, args.o)
def load_rules(path, *, encoding='utf-8'):
"""Load parse file with 'pseudo' tree of rules.
Args:
path: string with a path to the file with rules,
encoding: encoding of the file (default='utf-8')
Return:
dict: dictionary with pseudo tree structure, representing
hierarchy of rules
NOTE:
Expected file structure:
+- rule1_from --> rule1_to
| +- sub_rule1_from --> sub_rule1_to
| | +- sub_sub_rule1_from --> sub_sub_rule1_to
+- rule2_from --> rule2_to
+- rule3_from --> rule3_to
| +- sub_rule3_from --> sub_rule3_to
"""
# Load rules from a file.
raw_rules = load_file(path, encoding=encoding)
# Parse rules into 'pseudo' tree structure.
return parse_rules(list(raw_rules))
def read_file(filename, n_fold):
data_sources = []
parts = []
part_of_data = utils.load_file(filename)
part_of_data = part_of_data[1:len(part_of_data)]
part_of_data = filter_content(part_of_data)
parts.append(part_of_data)
parts = [item for sublist in parts for item in sublist]
data_sources = array(parts)
random.shuffle(data_sources)
data_count_limit = len(data_sources) / n_fold
folds = []
count = 0
for begin_index in range(0, len(data_sources), data_count_limit):
end_index = begin_index + data_count_limit
if (count == n_fold - 1):
end_index = len(data_sources)
print("begin: " + str(begin_index))
print("end: " + str(end_index))
folds.append(data_sources[begin_index:end_index])
count += 1
return folds
def read_data(args):
data_sources = []
header = ''
if (args.f != None):
parts = []
afile = args.f
part_of_data = utils.load_file(afile)
if args.e != None and args.e == 'y':
if header == '':
header = part_of_data[0]
part_of_data = part_of_data[1:len(part_of_data)]
part_of_data = calculat_hsv_model(part_of_data, args.t)
parts.append(part_of_data)
parts = [item for sublist in parts for item in sublist]
data_sources = array(parts)
output = []
if header != '':
output.append(header)
for item in data_sources:
output.append(item)
return output
def main():
parser = argparse.ArgumentParser(description = 'Generate HITs for Amazon Mechnical Turk workers.')
parser.add_argument('-f', help = 'The mtk data source file.')
parser.add_argument('-o', help = 'The output file of used data.')
args = parser.parse_args()
data_sources = []
if (args.f != None):
data_sources = utils.load_file(args.f)
random.shuffle(data_sources)
db_collections = hit.setup_mongodb()
data_metainfo = hit.regex_datasource(data_sources)
images_metainfo = hit.query_imagedata_from_db(db_collections, data_metainfo)
# data_labels: flickr high interesting 1, flickr low interesting 2, pinterest [3, 4, 5]
data_labels = data_metainfo[0]
# data_ids: (flickr, pinterest) image id
data_ids = data_metainfo[1]
data_count_limit = 50
for begin_index in range(0, len(data_sources), data_count_limit):
print("index: " + str(begin_index))
generate_hits(data_sources[begin_index:begin_index + data_count_limit], begin_index, args, data_ids[begin_index:begin_index + data_count_limit], images_metainfo)
sys.exit(0)
def main():
parser = argparse.ArgumentParser(description = 'Generate HITs for Amazon Mechnical Turk workers.')
parser.add_argument('-f', help = 'The mtk data source file.')
parser.add_argument('-o', help = 'The output file of used data.')
parser.add_argument('-m', default = 'normal', help = 'The running mode in {normal, qua_init, qua}.')
parser.add_argument('-q', help = 'The qualification type id.')
parser.add_argument('-t', default = 'sandbox', help = 'The type of Mechanical Turk.')
args = parser.parse_args()
if (args.m == 'qua' and args.q == None):
print('Please give qualification type id if running in qualification mode.')
sys.exit(0)
data_sources = []
if (args.f != None):
data_sources = utils.load_file(args.f)
if (args.m != 'qua'):
random.shuffle(data_sources)
data_count_limit = 100
for begin_index in range(0, len(data_sources), data_count_limit):
print("index: " + str(begin_index))
generate_hits(args.t, data_sources[begin_index:begin_index + data_count_limit], begin_index, args)
sys.exit(0)
def welcome(update, context):
chat_id = update.effective_chat.id
for new_member in update.message.new_chat_members:
chat_title = update.message.chat.title
if not new_member.is_bot:
first_name = new_member.first_name
last_name = new_member.last_name
username = new_member.username
if last_name is None:
last_name = ""
message = WELCOME_NEW_MEMBER_MESSAGE.format(
first_name=first_name,
last_name=last_name,
username=username,
chat_title=chat_title,
)
context.bot.sendPhoto(
chat_id=chat_id,
photo=load_file(
context=context,
pickle_file=DEFAULT_PICKEL_FILE_PHOTO,
default_photo=DEFAULT_PHOTO_WELCOME,
chat_id=chat_id,
),
caption=message,
)
elif new_member.full_name == "PyLadies Brasil Bot":
context.bot.send_message(
chat_id=chat_id,
text=HELLO_MESSAGE.format(chat_title=chat_title))
def add_hits(filename, all_hits):
hits = utils.load_file(filename)
hits = hits[0:len(hits)]
all_hits.append(hits)
return all_hits
def read_data(args):
x_column = int(args.x)
y_column = int(args.y)
bin_length = int(args.b)
threshold = float(args.t)
x_threshold = int(args.d)
data_sources = []
if (args.f != None):
parts = []
afile = args.f
part_of_data = utils.load_file(afile)
part_of_data = calculat_hsv_figure(part_of_data, x_column, y_column, bin_length, threshold, x_threshold)
parts.append(part_of_data)
parts = [item for sublist in parts for item in sublist]
data_sources = array(parts)
output = []
for item in data_sources:
output.append(item)
return output
def main(args):
content_generator = load_file(args.transcript, encoding=args.encoding)
rules = load_rules(args.rules, encoding=args.encoding)
mapped = list(do_mapping(content_generator, rules))
formatted = format_data(mapped)
save_file(args.output, formatted, encoding=args.encoding)
def load_csv(path, encoding='utf-8'):
for item in load_file(path, encoding=encoding):
splitted = item.split(';')
# Change to SINGLE word file format
splitted[2] = splitted[2][:-3]
splitted[3] = splitted[3][:-3]
yield splitted
def main(args):
content_generator = load_file(args.transcript, encoding=args.encoding)
rules = load_rules(args.rules, encoding=args.encoding)
mapped = do_mapping(content_generator, rules)
cleaned = clean(mapped)
formatted = mlf_format_data(cleaned)
save_file(args.output, formatted, encoding=args.encoding)
def load(cls):
"""
Load rules from config file.
"""
if cls._loaded_rules is None:
log("Loading contextual rules...", "CYAN", True)
lx = load_file("corpus/contextual_rules.rls")
cls._loaded_rules = [r for r in lx.split(u"\n") if len(r) > 1]
return cls._loaded_rules
def load(cls):
if cls._loaded_rules is None:
log("Loading lemmatizer rules...", "CYAN", True)
lx = load_file("corpus/lemmatizer_rules.rls")
cls._loaded_rules = []
for line in lx.split(u"\n"):
els = line.split(u"\t")
if els[0] != u"":
cls._loaded_rules.append(els[0])
return cls._loaded_rules
def main(args):
# Load alignment of the phonemes
alignment = load_mlf_to_dict(args.mlf, clean=False)
# Merge items in the mlf if the output is from AP decoder
if (args.a):
alignment = process_ap(alignment)
# Load path to the processed WAV files, should corresponds to the alignment
paths = load_file(args.scp)
# Load WAV files
waves = get_wave(paths)
# Use Noiser library to the changing of the tempo
process(waves, alignment, args.output, int(args.tempo), args.skip)
def setUp(self):
super(TestPublish, self).setUp()
self.user = User('*****@*****.**')
self.login('*****@*****.**')
self.issue = models.Issue(subject='test')
self.issue.local_base = False
self.issue.put()
self.ps = models.PatchSet(parent=self.issue, issue=self.issue)
self.ps.data = load_file('ps1.diff')
self.ps.save()
self.patches = engine.ParsePatchSet(self.ps)
db.put(self.patches)
def setUp(self):
super(TestStatusListener, self).setUp()
self.user = users.User('*****@*****.**')
self.login('*****@*****.**')
self.issue = models.Issue(subject='test')
self.issue.local_base = False
self.issue.put()
self.ps = models.PatchSet(parent=self.issue.key, issue_key=self.issue.key)
self.ps.data = load_file('ps1.diff')
self.ps.put()
self.patches = engine.ParsePatchSet(self.ps)
ndb.put_multi(self.patches)
self.logout() # Need to log out for /status_listener to work
def read_data(args):
data_sources = []
if (args.f != None):
if not isinstance(args.f, basestring):
parts = []
for afile in args.f:
part_of_data = utils.load_file(afile)
parts.append(part_of_data.tolist())
parts = [item for sublist in parts for item in sublist]
data_sources = array(parts)
else:
data_sources = utils.load_file(args.f)
hits = []
head_of_hits = ''
if (args.m != None):
if not isinstance(args.f, basestring):
parts = []
for afile in args.m:
part_of_data = utils.load_file(afile)
if head_of_hits == '':
head_of_hits = part_of_data[0]
part_of_data = part_of_data[1:len(part_of_data)]
parts.append(part_of_data.tolist())
parts = [item for sublist in parts for item in sublist]
hits = array(parts)
else:
hits = utils.load_file(args.m)
hits = hits[0]
hits = hits[1:len(hits)]
return (data_sources, hits, head_of_hits)
def main():
parser = argparse.ArgumentParser(description = 'Generate HITs for Amazon Mechnical Turk workers.')
parser.add_argument('-f', help = 'The mtk data source file.')
parser.add_argument('-o', help = 'The output file of used data.')
args = parser.parse_args()
data_sources = []
if (args.f != None):
data_sources = utils.load_file(args.f)
random.shuffle(data_sources)
generate_hits(data_sources, args)
sys.exit(0)
def loaded(self):
"""
Load lexicon in RAM, from file.
The representation will be a dict {"word1": [{tag1 : lemme1}]}
"""
if not self.PATH in self._loaded: # Caching and lazy loading
sulci_logger.debug("Loading lexicon...", "RED", True)
lx = load_file("%s/lexicon.lxc" % self.PATH)
self._loaded[self.PATH] = {}
for line in lx.split("\n"):
if line:
lexicon_entity = LexiconEntity(line)
self.add_factors(lexicon_entity.word)
self._loaded[self.PATH][lexicon_entity.word] = lexicon_entity
return self._loaded[self.PATH]
def read_file(filename):
data_sources = []
parts = []
part_of_data = utils.load_file(filename)
part_of_data = part_of_data[1 : len(part_of_data)]
part_of_data = filter_content(part_of_data)
parts.append(part_of_data)
parts = [item for sublist in parts for item in sublist]
dictionary = {}
for item in parts:
dictionary[item] = 1
return dictionary