Steboss

# lowest subsample
scale_low = 5 
# highest subsample
scale_high = 9 
# intervals between samples
scale_dense = 0.25 
# create an array of scales
scales = (2**np.arange(scale_low, scale_high, scale_dense)).astype(np.intc)
# ...
# here there will be a cycle and define what is the current scale

Steboss

avg_x = np.mean(X) 
Xt = [] 
for i, val in enumerate(X, 0):
    if i==0:
        Xt.append(val - avg_x)
    else:
        Xt.append(Xt[i-1] + (val - avg_x))

Steboss

Steboss

, June-July Biden, June-July Trump, September Biden, September Trump
# pro comments, 34'096, 61'525, 18'459, 45'670
# contra comments,27,817, 57'820, 29'844, 61'035

Steboss

from keras.preprocessing.text import Tokenizer
from keras.preprocessing.sequence import pad_sequences
from keras.models import Sequential
from keras.layers import LSTM, Dense, Bidirectional
from sklearn.model_selection import train_test_split 


# Convert text to array with Glove 
def text_to_array(text):
    r"""function to convert the text to vector through Glove"""

Steboss

from sklearn.feature_extraction.text import TfidfVectorizer,CountVectorizer
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import f1_score, roc_auc_score
from sklearn.model_selection import KFold

word_vectorizer = TfidfVectorizer(
    ngram_range=(1,4),
    min_df=3,
    max_df=0.9,
    use_idf=True,

Steboss

word_vectorizer = TfidfVectorizer(
    ngram_range=(1,4),
    min_df=3,
    max_df=0.9,
    use_idf=True,
    smooth_idf=True,
    sublinear_tf=True,
    analyzer='word',
    token_pattern=r'\w{1,}',
    max_features=50000)

Steboss

Steboss

from distutils.core import setup
from distutils.extension import Extension
from Cython.Build import cythonize

libraries = [ "fftw3","m"]
extra_compile_args = ['-O3', '-std=c99']
extra_link_args = ['-O3', '-std=c99']

cython_directives = {
    'embedsignature': True,

Steboss

#translate the pointer to a numpy array 
magnitude_array = np.asarray(magnitude)

#reshape the magnitude_array so we'll have a matrix
cols = int( (length/(windowSize/2)) -1)
rows = int(windowSize/2)+1

new_array = np.zeros([cols,rows])
counter = 0
for i in range(0,cols):