harshitcodes

CREATE OR REPLACE TABLE ecommerce_analytics.subset_user_sessions
OPTIONS(
 description="Obfuscated data from Google merchandise store"
) AS
SELECT * FROM `next-marketing-analytics.ecommerce.all_sessions_raw`
WHERE date = '20170101';

harshitcodes

class NeuralNet:
    def __init__(self, x, y):
        self.input      = x
        self.y          = y
        self.weights1   = np.random.rand(self.input.shape[1],4) 
        self.weights2   = np.random.rand(4,1)
        self.output     = np.zeros(y.shape)

harshitcodes

ran_int = np.random.randint(0, 10000, size=(rows, 2))

# rows to be written one by one
for i in range(rows):
  pointer['Num1'] = ran_int[i, 0]
  pointer['Num2'] = ran_int[i, 1]
  pointer.append()
  # this appends the data and
  # moves the pointer one row forward
tab.flush()

harshitcodes

filename = '<path to your folder>' + 'tab.h5'
h5 = tb.open_file(filename, 'w')

# creating 2million rows for the database
rows = 2000000
table_model = {
'Num1': tb.IntCol(pos=1),
'Num2': tb.IntCol(pos=2)
}
filters = tb.Filters(complevel=0) # no compression 

harshitcodes

import numpy as np
from random import gauss
path = '<enter the path where you want to write the file>'

a1 = [gauss(1.5, 2) for i in range(1000000)]
import pickle
pkl_file = open(path + "serialized_data.pkl", 'wb')

%time pickle.dump(a1, pkl_file)

harshitcodes

def record_comp_time(func_list, data_list, rep=3, number=1):
    ''' Function to compare the performance of different functions.
    Args:
    func_list : list
      list with function names as strings
    data_list : list
      list with data set names as strings
    rep : int
        number of repetitions of the whole comparison
    number : int

harshitcodes

plt.subplots(figsize=(12,10))
list1=[]

# extending the list of genres to collect all the genres of all the profitable movies
for i in profit_data['genres']:
    list1.extend(i)

    
genre_count_series = pd.Series(list1).value_counts()[:10].sort_values(ascending=True)
# output looks like

harshitcodes

# formatting the data in the genres columns.
movies_df['genres']=movies_df['genres'].str.strip('[]').str.replace(' ','').str.replace("'",'')
movies_df['genres']=movies_df['genres'].str.split(',')
movies_df.head()

harshitcodes

#selecting the movies having profit $50M or more
profit_data = movies_df[movies_df['profit'] >= 50000000]

#reindexing new data
profit_data.index = range(len(profit_data))

#we will start from 1 instead of 0
profit_data.index = profit_data.index + 1

harshitcodes

# Most profitable year from the given dataset.

profits_year.idxmax()