#file structure
!pip3 install cookiecutter

#general
!pip3 install --upgrade pip
!pip3 install ipython-autotime --quiet
!pip3 install watermark

%load_ext autotime

#get libraries
import pandas as pd
import os
import numpy as np
import itertools
from itertools import combinations
import warnings

#data wrangling
from functools import reduce

#get visualization libraries
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import seaborn as sns
import matplotlib.patches as mpatches
from matplotlib.ticker import FuncFormatter
import matplotlib.ticker as mtick
from matplotlib.ticker import PercentFormatter
from matplotlib.patches import Patch

#ML preprocessing
from sklearn.preprocessing import StandardScaler

#get ML functions
from sklearn.model_selection import train_test_split, cross_validate, cross_val_score, GridSearchCV, StratifiedKFold, learning_curve
from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier
from sklearn import __version__ as sklearn_version
import datetime

#get ML metric functions
from sklearn.metrics import accuracy_score, precision_score, recall_score, auc, precision_recall_curve, confusion_matrix

#get icr module and associated common functions
import in_vehicle_coupon_recommendation as icr
from in_vehicle_coupon_recommendation import p, rcp, rpp, rcr, pl, pdc, save_and_return_data_frame, initialize_custom_notebook_settings

#initialize notebook
st='no'; number_of_replicates=100 if st=='yes' else 10000

filename_version='4dot3'

initialize_custom_notebook_settings()

%load_ext autoreload
%autoreload 1
%aimport in_vehicle_coupon_recommendation 
%load_ext watermark

df = pd.read_csv(os.path.join('..', 'data', 'raw', 'in-vehicle-coupon-recommendation.csv'))
df = df.sample(frac=1, random_state=200) #row shuffle DataFrame
#p(df)

#rename 'passanger' and 'coupon' column
df = df.rename(columns={'passanger':'passenger', 'coupon':'coupon_venue_type'})

#decode string 'age' values to (explicit) string age ranges
column_name_age_value_list = list(df.loc[:, 'age'].unique())
column_name_age_value_list_cleaned_up = ['21-25', '46-49', '26-30', '31-35', '41-45', '50+', '36-40', '<21']
column_name_age_value_dict = dict(zip(column_name_age_value_list, column_name_age_value_list_cleaned_up))
df.loc[:, 'age'] = df.loc[:, 'age'].replace(column_name_age_value_dict)

del column_name_age_value_dict, column_name_age_value_list, column_name_age_value_list_cleaned_up


#drop column 'toCoupon_GEQ5min' because all 1's
column_name_list_not_toCoupon_GEQ5min = [column_name for column_name in df.columns if column_name != 'toCoupon_GEQ5min']
df = df.loc[:, column_name_list_not_toCoupon_GEQ5min]

del column_name_list_not_toCoupon_GEQ5min


#rename values in columns 'Bar', 'CoffeeHouse', 'CarryAway', 'RestaurantLessThan20', and 'Restaurant20To50' to math language...
#get column name list for column values uniques that are the same as column name 'CarryAway' value uniques
column_name_list_same_unique_values = []
for column_name in df.columns:    
    value = icr.column_name_value_sets_equal(df, column_name1='CarryAway', column_name2=column_name)
    if value == 1:
        column_name_list_same_unique_values = column_name_list_same_unique_values + [column_name]
#rename values in columns 'Bar', 'CoffeeHouse', 'CarryAway', 'RestaurantLessThan20', and 'Restaurant20To50' to math language
column_name_Bar_value_list = list(df.loc[:, 'Bar'].unique())
column_name_Bar_value_list_cleaned_up = ['never', '<1', '1-3', '>8', np.nan, '4-8']
column_name_Bar_value_dict = dict(zip(column_name_Bar_value_list, column_name_Bar_value_list_cleaned_up))
df.loc[:, column_name_list_same_unique_values] = df.loc[:, column_name_list_same_unique_values].replace(column_name_Bar_value_dict)

del column_name_Bar_value_dict, column_name_Bar_value_list, column_name_Bar_value_list_cleaned_up, column_name_list_same_unique_values


#clean up columns 'direction_same' and 'direction_opp' to 'direction_same_or_opposite'
#drop column 'direction_same'
column_name_list_not_direction_same = [column_name for column_name in df.columns if column_name != 'direction_same']
df = df.loc[:, column_name_list_not_direction_same]

del column_name_list_not_direction_same

#rename 'direction_opp' to 'direction_same_or_opposite'
df = df.rename(columns={'direction_opp':'direction_same_or_opposite'})


#solution1: fill in missing values with 'no response' or 'unknown'
column_name_list = ['Bar', 'CoffeeHouse', 'CarryAway', 'RestaurantLessThan20', 'Restaurant20To50', 'car']
df.loc[:, column_name_list] = df.loc[:, column_name_list].fillna('no response')


#fix income values
income_list = df.loc[:, 'income'].drop_duplicates().to_list()
income_list_fixed = ['\$12500 - \$24999', '\$87500 - \$99999', '\$37500 - \$49999', '\$50000 - \$62499', '\$75000 - \$87499', '\$25000 - \$37499', '\$100000 or More', 'Less than \$12500', '\$62500 - \$74999']
income_dictionary = dict(zip(income_list, income_list_fixed))
df.loc[:, 'income'] = df.loc[:, 'income'].replace(income_dictionary)
p(df)

#takeaway: the most represented scenario is heading to Work, alone, sunny weather, 55 degrees, 7am, bar, 1 day til expiration, >25 minute away in the opposite drive direction of destination
column_name_list_scenario = ['destination', 'passenger', 'weather', 'temperature', 'time', 'coupon_venue_type', 'expiration', 'toCoupon_GEQ15min', 'toCoupon_GEQ25min', 'direction_same_or_opposite']
print(df.loc[:, column_name_list_scenario].drop_duplicates().shape)
df.loc[:, column_name_list_scenario].value_counts().to_frame().reset_index().rename(columns={0:'Count'}).head(10)

#takeaway: the most represented demographic is Female, 31-35, Married partner, 0, Some college - no degree	Computer & Mathematical, Income $100000 or More, car no response, Bar never, CoffeeHouse never, Takeout no response, low-cost restaurant >8, mid-range restaurant never
column_name_list_not_demographic=['destination', 'passenger', 'weather', 'temperature', 'time', 'coupon_venue_type', 'expiration', 'toCoupon_GEQ15min', 'toCoupon_GEQ25min', 'direction_same_or_opposite', 'Y',]
column_name_list_demographic=[column_name for column_name in df.columns if not column_name in column_name_list_not_demographic]
df.loc[:, column_name_list_demographic].value_counts().reset_index().rename(columns={0:'Count'}).head(10)

#category representative numeric encoding: expiration, time, age, income

expiration_category_representative_numeric_encoding={'expiration':{'2h':2, '1d':24}}
time_category_representative_numeric_encoding={'time':{'7AM':7,'10AM':10, '2PM':14, '6PM':18, '10PM':22}}
age_category_representative_numeric_encoding={'age':{'<21':18, '21-25':23, '26-30':28, '31-35':33, '36-40':38, '41-45':43, '46-49':48, '50+':56}}
income_category_representative_numeric_encoding={'income':{'Less than \$12500':6250, '\$12500 - \$24999':18749.5, '\$25000 - \$37499':31249.5, '\$37500 - \$49999':43749.5, '\$50000 - \$62499':56249.5, '\$62500 - \$74999':68749.5, '\$75000 - \$87499':81249.5, '\$87500 - \$99999':93749.5,'\$100000 or More':150000}}

#combine category representative numeric encoding dictionaries
category_representative_numeric_encoding_dictionary = expiration_category_representative_numeric_encoding|time_category_representative_numeric_encoding|age_category_representative_numeric_encoding|income_category_representative_numeric_encoding

del expiration_category_representative_numeric_encoding, time_category_representative_numeric_encoding, age_category_representative_numeric_encoding, income_category_representative_numeric_encoding

category_representative_numeric_encoding_key_list = list(category_representative_numeric_encoding_dictionary.keys())

df_category_representative_numeric_encoding = df.loc[:, category_representative_numeric_encoding_key_list].replace(category_representative_numeric_encoding_dictionary)

del category_representative_numeric_encoding_dictionary, 

#rename columns
category_representative_numeric_encoding_substring='_category_representative_numeric_encoding'

column_name_list_category_representative_numeric_encoding = [str(column_name) + category_representative_numeric_encoding_substring for column_name in category_representative_numeric_encoding_key_list]

column_name_dictionary_category_representative_numeric_encoding = dict(zip(category_representative_numeric_encoding_key_list, column_name_list_category_representative_numeric_encoding))

df_category_representative_numeric_encoding = df_category_representative_numeric_encoding.rename(columns=column_name_dictionary_category_representative_numeric_encoding)

del column_name_dictionary_category_representative_numeric_encoding, column_name_list_category_representative_numeric_encoding, category_representative_numeric_encoding_key_list, category_representative_numeric_encoding_substring

#p(df_category_representative_numeric_encoding)

#binary encoding: gender, expiration
gender_binary_encoding={'gender':{'Female':0, 'Male':1}}
expiration_binary_encoding={'expiration':{'2h':0, '1d':1}}

binary_encoding_dictionary = gender_binary_encoding|expiration_binary_encoding

del gender_binary_encoding, expiration_binary_encoding

binary_encoding_key_list = list(binary_encoding_dictionary.keys())

df_binary_encoding = df.loc[:, binary_encoding_key_list].replace(binary_encoding_dictionary)

del binary_encoding_dictionary


#rename columns
binary_encoding_substring = '_binary_encoding'
column_name_list_binary_encoding = [str(column_name) + binary_encoding_substring for column_name in binary_encoding_key_list]
column_name_dictionary_binary_encoding = dict(zip(binary_encoding_key_list, column_name_list_binary_encoding))
df_binary_encoding = df_binary_encoding.rename(columns=column_name_dictionary_binary_encoding)

del column_name_dictionary_binary_encoding, column_name_list_binary_encoding, binary_encoding_key_list, binary_encoding_substring

#p(df_binary_encoding)

#ordinal integer encoding: coupon_venue_type, education, income, age, time, temperature

#category ordinal features to ordinal integer encoding
coupon_venue_type_ordinal_integer_encoding={'coupon_venue_type':{'Coffee House':1, 'Bar':2, 'Carry out & Take away':3, 'Restaurant(<20)':4, 'Restaurant(20-50)':5}}
education_ordinal_integer_encoding={'education':{'Some High School':1, 'High School Graduate':2, 'Some college - no degree':3, 'Associates degree':4, 'Bachelors degree':5,'Graduate degree (Masters or Doctorate)':6}}
income_ordinal_integer_encoding={'income':{'Less than \$12500':1, '\$12500 - \$24999':2, '\$25000 - \$37499':3, '\$37500 - \$49999':4, '\$50000 - \$62499':5, '\$62500 - \$74999':6, '\$75000 - \$87499':7, '\$87500 - \$99999':8, '\$100000 or More':9}}
age_ordinal_integer_encoding={'age':{'<21':1, '21-25':2, '26-30':3, '31-35':4, '36-40':5, '41-45':6, '46-49':7, '50+':8}}
time_ordinal_integer_encoding={'time':{'7AM':1, '10AM':2, '2PM':3, '6PM':4, '10PM':5}}

#numeric to ordinal integer encoding
temperature_ordinal_integer_encoding={'temperature':{30:1, 55:2, 80:3}}


ordinal_integer_encoding_dictionary = coupon_venue_type_ordinal_integer_encoding|education_ordinal_integer_encoding|income_ordinal_integer_encoding|age_ordinal_integer_encoding|time_ordinal_integer_encoding|temperature_ordinal_integer_encoding

del coupon_venue_type_ordinal_integer_encoding, education_ordinal_integer_encoding, income_ordinal_integer_encoding, age_ordinal_integer_encoding, time_ordinal_integer_encoding, temperature_ordinal_integer_encoding

ordinal_integer_encoding_key_list = list(ordinal_integer_encoding_dictionary.keys())

df_ordinal_integer_encoding = df.loc[:, ordinal_integer_encoding_key_list].replace(ordinal_integer_encoding_dictionary)

del ordinal_integer_encoding_dictionary

#rename columns
ordinal_integer_encoding_substring='_ordinal_integer_encoding'
column_name_list_ordinal_integer_encoding=[str(column_name) + ordinal_integer_encoding_substring for column_name in ordinal_integer_encoding_key_list]
column_name_dictionary_ordinal_integer_encoding=dict(zip(ordinal_integer_encoding_key_list,column_name_list_ordinal_integer_encoding))
df_ordinal_integer_encoding=df_ordinal_integer_encoding.rename(columns=column_name_dictionary_ordinal_integer_encoding)

del column_name_dictionary_ordinal_integer_encoding, column_name_list_ordinal_integer_encoding, ordinal_integer_encoding_key_list, ordinal_integer_encoding_substring

#p(df_ordinal_integer_encoding)

#venue type visits per month yes response to ordinal integer encoding: bar, coffeehouse, carryaway, restaurantlessthan20, restaurant20to50
bar_venue_visit_frequency_yes_response_ordinal_integer_encoding={'Bar':{'never':1, '<1':2, '1-3':3, '4-8':4, '>8':5, 'no response':0}}
coffeehouse_venue_visit_frequency_yes_response_ordinal_integer_encoding={'CoffeeHouse':{'never':1, '<1':2, '1-3':3, '4-8':4, '>8':5, 'no response':0}}
carryaway_venue_visit_frequency_yes_response_ordinal_integer_encoding={'CarryAway':{'never':1, '<1':2, '1-3':3, '4-8':4, '>8':5, 'no response':0}}
restaurantlessthan20_venue_visit_frequency_yes_response_ordinal_integer_encoding={'RestaurantLessThan20':{'never':1, '<1':2, '1-3':3, '4-8':4, '>8':5, 'no response':0}}
restaurant20to50_venue_visit_frequency_yes_response_ordinal_integer_encoding={'Restaurant20To50':{'never':1, '<1':2, '1-3':3, '4-8':4, '>8':5, 'no response':0}}


venue_visit_frequency_yes_response_ordinal_integer_encoding_dictionary=bar_venue_visit_frequency_yes_response_ordinal_integer_encoding|coffeehouse_venue_visit_frequency_yes_response_ordinal_integer_encoding|carryaway_venue_visit_frequency_yes_response_ordinal_integer_encoding|restaurantlessthan20_venue_visit_frequency_yes_response_ordinal_integer_encoding|restaurant20to50_venue_visit_frequency_yes_response_ordinal_integer_encoding

del bar_venue_visit_frequency_yes_response_ordinal_integer_encoding, carryaway_venue_visit_frequency_yes_response_ordinal_integer_encoding, restaurantlessthan20_venue_visit_frequency_yes_response_ordinal_integer_encoding, restaurant20to50_venue_visit_frequency_yes_response_ordinal_integer_encoding
#del bar_venue_visit_frequency_yes_response_ordinal_integer_encoding, coffeehouse_venue_visit_frequency_yes_response_ordinal_integer_encoding, carryaway_venue_visit_frequency_yes_response_ordinal_integer_encoding, restaurantlessthan20_venue_visit_frequency_yes_response_ordinal_integer_encoding, restaurant20to50_venue_visit_frequency_yes_response_ordinal_integer_encoding

venue_visit_frequency_yes_response_ordinal_integer_encoding_key_list=list(venue_visit_frequency_yes_response_ordinal_integer_encoding_dictionary.keys())

df_venue_visit_frequency_yes_response_ordinal_integer_encoding = df.loc[:, venue_visit_frequency_yes_response_ordinal_integer_encoding_key_list].replace(venue_visit_frequency_yes_response_ordinal_integer_encoding_dictionary)

del venue_visit_frequency_yes_response_ordinal_integer_encoding_dictionary

#rename columns
venue_visit_frequency_yes_response_ordinal_integer_encoding_substring='_venue_visit_frequency_yes_response_ordinal_integer_encoding'
column_name_list_venue_visit_frequency_yes_response_ordinal_integer_encoding=[str(column_name) + venue_visit_frequency_yes_response_ordinal_integer_encoding_substring for column_name in venue_visit_frequency_yes_response_ordinal_integer_encoding_key_list]
column_name_dictionary_venue_visit_frequency_yes_response_ordinal_integer_encoding=dict(zip(venue_visit_frequency_yes_response_ordinal_integer_encoding_key_list, column_name_list_venue_visit_frequency_yes_response_ordinal_integer_encoding))
df_venue_visit_frequency_yes_response_ordinal_integer_encoding = df_venue_visit_frequency_yes_response_ordinal_integer_encoding.rename(columns=column_name_dictionary_venue_visit_frequency_yes_response_ordinal_integer_encoding)

del column_name_dictionary_venue_visit_frequency_yes_response_ordinal_integer_encoding, venue_visit_frequency_yes_response_ordinal_integer_encoding_key_list, venue_visit_frequency_yes_response_ordinal_integer_encoding_substring, column_name_list_venue_visit_frequency_yes_response_ordinal_integer_encoding

#p(df_venue_visit_frequency_yes_response_ordinal_integer_encoding)

#venue type visits per month no response to indicator variable: bar, coffeehouse, carryaway, restaurantlessthan20, restaurant20to50
bar_venue_visit_frequency_no_response_indicator_variable={'Bar':{'never':0, '<1':0, '1-3':0, '4-8':0, '>8':0, 'no response':1}}
coffeehouse_venue_visit_frequency_no_response_indicator_variable={'CoffeeHouse':{'never':0, '<1':0, '1-3':0, '4-8':0, '>8':0, 'no response':1}}
carryaway_venue_visit_frequency_no_response_indicator_variable={'CarryAway':{'never':0, '<1':0, '1-3':0, '4-8':0, '>8':0, 'no response':1}}
restaurantlessthan20_venue_visit_frequency_no_response_indicator_variable={'RestaurantLessThan20':{'never':0, '<1':0, '1-3':0, '4-8':0, '>8':0, 'no response':1}}
restaurant20to50_venue_visit_frequency_no_response_indicator_variable={'Restaurant20To50':{'never':0, '<1':0, '1-3':0, '4-8':0, '>8':0, 'no response':1}}

venue_visit_frequency_no_response_indicator_variable_dictionary=bar_venue_visit_frequency_no_response_indicator_variable|coffeehouse_venue_visit_frequency_no_response_indicator_variable|carryaway_venue_visit_frequency_no_response_indicator_variable|restaurantlessthan20_venue_visit_frequency_no_response_indicator_variable|restaurant20to50_venue_visit_frequency_no_response_indicator_variable

del bar_venue_visit_frequency_no_response_indicator_variable, coffeehouse_venue_visit_frequency_no_response_indicator_variable, carryaway_venue_visit_frequency_no_response_indicator_variable, restaurantlessthan20_venue_visit_frequency_no_response_indicator_variable, restaurant20to50_venue_visit_frequency_no_response_indicator_variable

venue_visit_frequency_no_response_indicator_variable_key_list=list(venue_visit_frequency_no_response_indicator_variable_dictionary.keys())

df_venue_visit_frequency_no_response_indicator_variable=df.loc[:, venue_visit_frequency_no_response_indicator_variable_key_list].replace(venue_visit_frequency_no_response_indicator_variable_dictionary)

del venue_visit_frequency_no_response_indicator_variable_dictionary

#rename columns
venue_visit_frequency_no_response_indicator_variable='_venue_visit_frequency_no_response_indicator'
column_name_list_venue_visit_frequency_no_response_indicator_variable=[str(column_name) + venue_visit_frequency_no_response_indicator_variable for column_name in venue_visit_frequency_no_response_indicator_variable_key_list]
column_name_dictionary_venue_visit_frequency_no_response_indicator_variable=dict(zip(venue_visit_frequency_no_response_indicator_variable_key_list, column_name_list_venue_visit_frequency_no_response_indicator_variable))
df_venue_visit_frequency_no_response_indicator_variable=df_venue_visit_frequency_no_response_indicator_variable.rename(columns=column_name_dictionary_venue_visit_frequency_no_response_indicator_variable)

del column_name_dictionary_venue_visit_frequency_no_response_indicator_variable, column_name_list_venue_visit_frequency_no_response_indicator_variable, venue_visit_frequency_no_response_indicator_variable_key_list, venue_visit_frequency_no_response_indicator_variable

#p(df_venue_visit_frequency_no_response_indicator_variable)

#concatenate the data frames

#get data frame list
data_frame_list=[df, df_category_representative_numeric_encoding, df_binary_encoding, df_ordinal_integer_encoding, df_venue_visit_frequency_yes_response_ordinal_integer_encoding, df_venue_visit_frequency_no_response_indicator_variable]
del df_category_representative_numeric_encoding, df_binary_encoding, df_ordinal_integer_encoding, df_venue_visit_frequency_yes_response_ordinal_integer_encoding, df_venue_visit_frequency_no_response_indicator_variable

#concat data frames
df = pd.concat(data_frame_list, axis=1)
del data_frame_list

p(df)

#save df collection with four object train test splits before preprocessed DataFrame
df_collection_filename='df_collection_train_test_v' + filename_version + '.pkl'


df_readback = icr.return_processed_collection_if_it_exists(filename=df_collection_filename)
if df_readback != None:
    df_collection = df_readback
else:
    df_collection = {}
    
    #split the data into train and test
    df_collection['X_train'], df_collection['X_test'], df_collection['Y_train'], df_collection['Y_test'] = \
    train_test_split(df.drop(columns=['Y']), df.loc[:, 'Y'], test_size=.2, random_state=200)
    
    #save preprocessed data frame collection
    df_collection = icr.save_and_return_collection(data_frame_collection=df_collection, filename=df_collection_filename)

p(df_collection['X_train'])

color_list = ['#a6cee3','#1f78b4','#b2df8a','#33a02c','#fb9a99','#e31a1c','#fdbf6f','#ff7f00','#cab2d6']

bar_orientation='vertical'
dpi=100
figure_filename = '../reports/figures/figure_coupon_acceptance_percentage_and_frequency_distribution_'+ str(bar_orientation) + '_bar_plot_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'
figure_filename_axes1 = '../reports/figures/figure_coupon_acceptance_percentage_and_frequency_distribution_'+ str(bar_orientation) + '_bar_plot_axes1_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'
figure_filename_axes2 = '../reports/figures/figure_coupon_acceptance_percentage_and_frequency_distribution_'+ str(bar_orientation) + '_bar_plot_axes2_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

plot_save_together_separate = 'together'

figsize=(12, 6.2)
bar_width=0.8

xlabel = 'Coupon Acceptance'
ylabel = 'Percentage'
title = 'Coupon Acceptance '+ str(ylabel) + ' Distribution'

target_value_dictionary = {1: 'Yes', 0:'No'}

df_Y_frequency = df.loc[:, ['Y']].value_counts().reset_index().rename(columns={0:'frequency'})
df_Y_frequency.loc[:, 'percentage'] = df_Y_frequency.loc[:, 'frequency'] / df_Y_frequency.loc[:, 'frequency'].sum(axis=0) * 100
df_Y_frequency_percentage = df_Y_frequency
del df_Y_frequency

x_label_list = df_Y_frequency_percentage.loc[:, 'Y'].replace(target_value_dictionary).to_list()


if plot_save_together_separate == 'together':
    figure, (axes1, axes2) = plt.subplots(nrows=1, ncols=2, figsize=figsize)
elif plot_save_together_separate == 'separate':
    figure1, axes1 = plt.subplots(nrows=1, ncols=1, figsize=figsize)
    figure2, axes2 = plt.subplots(nrows=1, ncols=1, figsize=figsize)
    

#make percentage subplot
annotation_round_by_number = 0

percentage_list = df_Y_frequency_percentage.loc[:, 'percentage'].to_list()

axes1.bar(x=x_label_list, height=percentage_list, width=bar_width, color=[color_list[3], color_list[0]])

axes1.set_title(label=title, fontsize=18)
axes1.set_xlabel(xlabel=xlabel, fontsize=17)
axes1.set_ylabel(ylabel=ylabel, fontsize=17)
axes1.tick_params(axis='both', which='major', labelsize=15)

for rectangle, annotation in zip(axes1.patches, percentage_list):
    axes1.text(rectangle.get_x() + rectangle.get_width() / 2, rectangle.get_height(), format(round(annotation, annotation_round_by_number), '.0f') + '%', ha="center", va="bottom", fontsize=15)

#make frequency subplot
ylabel = 'Frequency'
title = 'Coupon Acceptance '+ str(ylabel) + ' Distribution'
annotation_round_by_number = -2

frequency_list = df_Y_frequency_percentage.loc[:, 'frequency'].to_list()

axes2.bar(x=x_label_list, height=frequency_list, width=bar_width, color=[color_list[3], color_list[0]])

axes2.set_title(label=title, fontsize=18)
axes2.set_xlabel(xlabel=xlabel, fontsize=17)
axes2.set_ylabel(ylabel=ylabel, fontsize=17)
axes2.tick_params(axis='both', which='major', labelsize=15)

for rectangle, annotation in zip(axes2.patches, frequency_list):
    axes2.text(rectangle.get_x() + rectangle.get_width() / 2, rectangle.get_height() + 5, round(annotation, annotation_round_by_number), ha="center", va="bottom", fontsize=15)

plt.tight_layout()

if plot_save_together_separate == 'together':
    figure.savefig(figure_filename, bbox_inches='tight', dpi=dpi)
elif plot_save_together_separate == 'separate':
    figure1.savefig(figure_filename_axes1, bbox_inches='tight', dpi=dpi)
    figure2.savefig(figure_filename_axes2, bbox_inches='tight', dpi=dpi)


plt.show()

#sample size, population size, margin of error, confidence interval

#intialize parameters
z_score = 1.96; confidence_interval = .95

p_value = 0.5684326710816777
margin_of_error = .0381
N_population_size = 2.4 * 10**8

#known 652 surveys acceptance, https://jmlr.org/papers/volume18/16-003/16-003.pdf



#calculate sample size
sample_size = (z_score**2*p_value*(1-p_value)/margin_of_error**2)/(1+(z_score**2*p_value*(1-p_value)/margin_of_error**2/N_population_size))
print('The '+ str(round(confidence_interval*100, None)) +'% confidence interval for a ' + str(round(sample_size)) + ' sample size has a ' \
      + str(margin_of_error) + ' margin of error and ' + str(round(p_value, 5)) + ' p-value representing a '+ '{:,}'.format(round(N_population_size)) +' population size'
      + '. The sample of 652 survey participants is slightly more than 649.\n')

round_by_number = None
print('Therefore, '+ str(round(confidence_interval*100, None)) +'% of the time, we expect a coupon acceptance rate between ' \
      + str(round((p_value-margin_of_error/2)*100, round_by_number)) + '% and ' \
      + str(round((p_value+margin_of_error/2)*100, round_by_number)) + '%')

feature_column_name = 'expiration'
feature_column_name_label=feature_column_name.capitalize()
xlabel=feature_column_name_label
ylabel='Frequency'

xtick_rotation=0
bar_orientation = 'vertical'
annotation_type = 'frequency'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['1d', '2h']
xtick_dictionary = {'1d':'1 day', '2h':'2 hours'}

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]


figsize=(8,8)


#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number)

feature_column_name = 'income'
feature_column_name_label=feature_column_name.capitalize()
xlabel=feature_column_name_label
ylabel='Percentage'

xtick_rotation=90
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2
rectangle_annotation_y_offset = -45

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['Less than \$12500', '\$12500 - \$24999', '\$25000 - \$37499', '\$37500 - \$49999', '\$50000 - \$62499', '\$62500 - \$74999',  '\$75000 - \$87499', '\$87500 - \$99999', '\$100000 or More',]
xtick_dictionary = None

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]


figsize=(8,8)


#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, rectangle_annotation_y_offset=rectangle_annotation_y_offset)

feature_column_name = 'direction_same_or_opposite'
feature_column_name_label=feature_column_name.capitalize()
xlabel=feature_column_name_label
ylabel='Frequency'

xtick_rotation=0
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = [0,1]
xtick_dictionary = {0:'Same', 1:'Opposite'}

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]

y_upper_limit=None

figsize=(8,8)


#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, y_upper_limit=y_upper_limit)

feature_column_name = 'expiration'
feature_column_name_label='Coupon '+feature_column_name.capitalize()
xlabel='Coupon '+feature_column_name
ylabel='Frequency'

xtick_rotation=0
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['1d', '2h']
xtick_dictionary = {'1d':'1 day', '2h':'2 hours'}

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]

y_upper_limit=None

figsize=(8,8)


#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, y_upper_limit=y_upper_limit)

feature_column_name = 'gender'
feature_column_name_label=feature_column_name.capitalize()
xlabel=feature_column_name_label
ylabel='Frequency'

xtick_rotation=0
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['Female','Male', ]
xtick_dictionary = None

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]

y_upper_limit=7500

figsize=(8,8)


#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, y_upper_limit=y_upper_limit)

feature_column_name = 'destination'
feature_column_name_label=feature_column_name.capitalize()
xlabel=feature_column_name_label
ylabel='Frequency'

xtick_rotation=0
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['No Urgent Place', 'Work', 'Home']
xtick_dictionary = {'No Urgent Place':'No urgent place', 'Work':'Work', 'Home':'Home'}

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]

y_upper_limit=None

figsize=(8,8)
rectangle_annotation_y_offset = -100


#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, y_upper_limit=y_upper_limit, rectangle_annotation_y_offset=rectangle_annotation_y_offset)

feature_column_name = 'passenger'
feature_column_name_label=feature_column_name.capitalize()
xlabel=feature_column_name_label
ylabel='Frequency'

xtick_rotation=0
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['Alone', 'Friend(s)', 'Partner', 'Kid(s)', ]
xtick_dictionary = None

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]

y_upper_limit=None

figsize=(8,8)

rectangle_annotation_y_offset=-140

#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, y_upper_limit=y_upper_limit, rectangle_annotation_y_offset=rectangle_annotation_y_offset)

feature_column_name = 'weather'
feature_column_name_label=feature_column_name.capitalize()
xlabel=feature_column_name_label
ylabel='Frequency'

xtick_rotation=0
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['Sunny', 'Snowy', 'Rainy', ]
xtick_dictionary = None

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]

y_upper_limit=None

figsize=(8,8)

rectangle_annotation_y_offset=-140

#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, y_upper_limit=y_upper_limit, rectangle_annotation_y_offset=rectangle_annotation_y_offset)

feature_column_name = 'time'
feature_column_name_label=feature_column_name.capitalize()
xlabel=feature_column_name_label
ylabel='Frequency'

xtick_rotation=0
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['7AM', '10AM', '2PM', '6PM', '10PM']
xtick_dictionary = None

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]

y_upper_limit=None

figsize=(8,8)

rectangle_annotation_y_offset=-80

#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, y_upper_limit=y_upper_limit, rectangle_annotation_y_offset=rectangle_annotation_y_offset)

feature_column_name = 'coupon_venue_type'
feature_column_name_label='Coupon Venue Type'#feature_column_name.capitalize()
xlabel='Coupon venue type'#feature_column_name_label
ylabel='Frequency'

xtick_rotation=90
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['Coffee House', 'Restaurant(<20)', 'Carry out & Take away', 'Bar', 'Restaurant(20-50)']
xtick_dictionary = {'Coffee House':'Coffee house', 'Restaurant(<20)':'Low-cost restaurant', 'Carry out & Take away':'Takeout', 'Bar':'Bar', 'Restaurant(20-50)':'Mid-range restaurant'}

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]

y_upper_limit=None

figsize=(8,6)

rectangle_annotation_y_offset=-80

#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, y_upper_limit=y_upper_limit, rectangle_annotation_y_offset=rectangle_annotation_y_offset, figsize=figsize)

feature_column_name = 'age'
feature_column_name_label='Age Group'#feature_column_name.capitalize()
xlabel='Age'#feature_column_name_label
ylabel='Frequency'

xtick_rotation=0
bar_orientation = 'vertical'
annotation_type = 'percentage'
frequency_annotation_round_by_number=-2

dpi=100
figure_filename = '../reports/figures/figure_'+ str(feature_column_name) +'_frequency_distribution_'+ str(bar_orientation) + '_stacked_bar_coupon_acceptance_coupon_refusal_'+ str(annotation_type) + '_annotation_all_data_dpi_' + str(dpi) + '_v' + filename_version + '.png'

feature_value_order_list = ['<21', '21-25', '26-30', '31-35', '36-40', '41-45', '46-49', '50+']
xtick_dictionary = None #{}

title= str(feature_column_name_label) + ' Frequency Distribution'

color_index_list=[3,0]
colors = [color_list[color_index_list[0]], color_list[color_index_list[1]]]

y_upper_limit=None

figsize=(8,6)

rectangle_annotation_y_offset=-40

#get column name, coupon acceptance (frequency), coupon refusal (frequency)
df_income_coupon_acceptance_coupon_refusal = icr.get_feature_target_frequency_data_frame(df, feature_column_name=feature_column_name, append_percentage_true_false=True)

#sort by value order list
df_income_coupon_acceptance_coupon_refusal = icr.sort_data_frame(df=df_income_coupon_acceptance_coupon_refusal, feature_column_name=feature_column_name, feature_value_order_list=feature_value_order_list, ascending_true_false=True).drop(columns=[str(feature_column_name)+'_rank'])

#index feature column
df_index_income_coupon_refusal_coupon_acceptance = df_income_coupon_acceptance_coupon_refusal.set_index(feature_column_name)

icr.plot_vertical_stacked_bar_graph(df=df_index_income_coupon_refusal_coupon_acceptance, feature_column_name=feature_column_name, feature_column_name_label=feature_column_name_label, ylabel=ylabel, xlabel=xlabel, figure_filename=figure_filename, annotation_text_size=14, dpi=dpi, xtick_rotation=xtick_rotation, xtick_dictionary=xtick_dictionary, colors=colors, annotation_type=annotation_type, frequency_annotation_round_by_number=frequency_annotation_round_by_number, y_upper_limit=y_upper_limit, rectangle_annotation_y_offset=rectangle_annotation_y_offset, figsize=figsize)