arseniyturin

model = Sequential()
model.add(Conv1D(filters=500, kernel_size=10, activation='relu', input_shape=(n_steps, n_features)))
model.add(MaxPooling1D(pool_size=10))
model.add(Flatten())
model.add(Dense(500, activation='relu'))
model.add(Dense(100, activation='relu'))
model.add(Dense(1, activation='sigmoid'))
model.compile(optimizer='adam', loss='mse')

arseniyturin

model = Sequential()
model.add(LSTM(100, activation='relu', return_sequences=True, input_shape=(n_steps, n_features)))
model.add(LSTM(50, activation='relu', return_sequences=False))
model.add(Dense(10))
model.add(Dense(1))
model.compile(optimizer='adam', loss='mse', metrics=['mae'])

arseniyturin

from math import factorial as fac

def binomial_prob(n, x, p):
    return round( ( fac(n) / ( fac(x) * fac(n-x) ) ) * (p**x) * (1-p)**(n-x) , 5)

arseniyturin

def SGD(X, y, lr=0.05, epoch=10, batch_size=1):
        
    '''
    Stochastic Gradient Descent for a single feature
    '''
    
    m, b = 0.5, 0.5 # initial parameters
    log, mse = [], [] # lists to store learning process
    
    for _ in range(epoch):

arseniyturin

from selenium import webdriver

driver = webdriver.Chrome('../chromedriver')
driver.get('https://www.instagram.com/accounts/login/')

def get_post_description():
    description = driver.execute_script('''return document.getElementsByClassName('C4VMK')[0].getElementsByTagName('span')[0].innerText''')
    return description 

def get_post_date():

arseniyturin

# emulate click on a likes button
driver.execute_script('''document.getElementsByClassName('Nm9Fw')[0].lastElementChild.click()''')

# scroll down by 4000px to load more users who liked the post
driver.execute_script('''document.getElementsByClassName('Igw0E IwRSH eGOV_ vwCYk i0EQd')[0].firstChild.scrollBy(0, 4000);''')
time.sleep(0.2)

# collect usernames
usernames = driver.execute_script('''return document.getElementsByClassName('_7UhW9 xLCgt MMzan KV-D4 fDxYl');''')

arseniyturin

from numpy import log, dot, e
from numpy.random import rand

class LogisticRegression:
    
    def sigmoid(self, z): return 1 / (1 + e**(-z))
    
    def cost_function(self, X, y, weights):                 
        z = dot(X, weights)
        predict_1 = y * log(self.sigmoid(z))

arseniyturin

/*

  Problem: 
  
  You have 16 bit number, for example 12345. 
  In hexadecimal it looks like this: 0x3039, 
  but in the buffer it will look like this: <Buffer: 39 30>
  
  How to convert two 8 bit numbers back to 16 bit number?
  Bitwise operator '<<'

arseniyturin

"""
Table 'Products' contains different products, their categories and prices.
We want to select top n products per each category rated by their price
SOLUTION: Window function RANK()

1. TABLE 'Products'
┌───────────┬────────────┬───────┐
│  product  │  category  │ price │
├───────────┼────────────┼───────┤
│ product_A │ category_A │ 55    │

arseniyturin

'''
    
            P(A) * (B|A)
  P(A|B) = -------------- - Bayes Probability
                P(B)   

  Cookie Problem that can be solved with Bayes Probability:
  
  You are given two jars of cookies, jar 1 has 30 vanilla and 10 chocolate cookies,
  jar 2 has 20 vanilla and 20 chocolate cookies.