# -*- coding: utf-8 -*-
"""

"""


import numpy as np
import matplotlib.pyplot as plt
import pandas as pd

from sklearn import preprocessing
from sklearn.model_selection import train_test_split
from sklearn import metrics
from sklearn.tree import DecisionTreeClassifier
from sklearn.neural_network import MLPClassifier

from sklearn.model_selection import KFold, StratifiedKFold
import numpy as np

import xgboost as xgb
from xgboost import plot_importance, plot_tree
from sklearn.ensemble import ExtraTreesRegressor
from sklearn.feature_selection import SelectFromModel



import seaborn as sns; sns.set()


import sys
if not sys.warnoptions:
    import warnings
    warnings.simplefilter("ignore")
#from sklearn.metrics import mean_absolute_percentage_error


import sys
if not sys.warnoptions:
    import warnings
    warnings.simplefilter("ignore")
    


   
if __name__ == '__main__':
    pd.options.display.max_columns=None
    df_ori = pd.read_csv('seoul hourly/2020_hourly.csv', sep = ',', encoding= 'unicode_escape')
    #df_ori = df_ori.dropna()
    #df_ori = df_ori[df_ori['Functioning Day'] != 'No']
    
    
    df_ori2 = df_ori[['hour', 'day_week', 'month', 'count']]
    df_ori2.columns = ['hour', 'day', 'month', 'rented_bikes']
    fig3 = plt.figure(figsize=(6,6))
    dt_plot = df_ori2
    g = sns.heatmap(dt_plot.corr(),cmap="BrBG",annot=False, center=0)
    dt_plot.corr()
    plt.xticks(fontsize=15)
    plt.yticks(fontsize=15)
    plt.tight_layout()
    plt.show()
    fig3.savefig('heat map.png', format='png', dpi=300)

    df_ori['date_time'] = pd.to_datetime(df_ori['date_time'])
    
    y = df_ori['count']
    x = df_ori['date_time'].dt.strftime('%Y-%m-%d %hh:%mm:%ss')
    interval = 500
    
    fig1=plt.figure(figsize=(14,6))
    #plt.title('Rented Bikes')
    plt.plot(x,y, label = '')
    plt.legend(loc='upper right', prop={'size':15})
    plt.xlabel('Date and time', fontsize=15)
    plt.ylabel('Number of rented bikes', fontsize=15)
    plt.gcf().autofmt_xdate()
    plt.xticks(np.arange(0, len(x)+1, interval))
    
    #plt.show()    
    fig1.savefig('rented bikes hourly.png', format='png', dpi=300)
    

    #------------------------
    df_day = df_ori.resample('d', on='date_time').mean().reset_index()
    
    y = df_day['count']
    x = df_day['date_time'].dt.strftime('%Y-%m-%d')
    interval = 14
    
    fig1=plt.figure(figsize=(14,5))
    #plt.title('Rented Bikes')
    plt.plot(x,y)
    plt.legend(loc='upper right', prop={'size':15})
    plt.xlabel('Date', fontsize=15)
    plt.ylabel('Average number of rented bikes', fontsize=15)
    plt.gcf().autofmt_xdate()
    plt.xticks(np.arange(0, len(x), interval))
    
    #plt.show()    
    fig1.savefig('rented bikes daily.png', format='png', dpi=300)
    
    
    #-----------------
    df_month = df_ori.resample('m', on='date_time').mean().reset_index()
    
    y = df_month['count']
    x = df_month['date_time'].dt.strftime('%Y-%m')
    interval = 1
    
    fig1=plt.figure(figsize=(14,4))
    #plt.title('Rented Bikes')
    plt.bar(x,y)
    plt.legend(loc='upper right', prop={'size':15})
    plt.xlabel('Month', fontsize=15)
    plt.ylabel('Average number of rented bikes', fontsize=15)
    plt.gcf().autofmt_xdate()
    plt.xticks(np.arange(0, len(x), interval))
    
    #plt.show()    
    fig1.savefig('rented bikes monthly.png', format='png', dpi=300)
    
    
    
    
    #------------------------
    df_hour = df_ori.groupby(['day_week','hour']).mean().reset_index()
    df_monday = df_hour[0:24]
    df_tuesday = df_hour[24:48]
    df_wednesday = df_hour[48:72]
    df_thursday =df_hour[72:96]
    df_friday = df_hour[96:120]
    df_saturday = df_hour[120:144]
    df_sunday = df_hour[144:168]
    
   

    fig1=plt.figure(figsize=(14,4))
    x = df_monday['hour']
    #plt.title('Rented Bikes')
    plt.plot(x, df_monday['count'].values, label = 'Monday')
    plt.plot(x, df_tuesday['count'].values, label = 'Tuesday')
    plt.plot(x, df_wednesday['count'].values, label = 'Wednesday')
    plt.plot(x, df_thursday['count'].values, label = 'Thursday')
    plt.plot(x, df_friday['count'].values, label = 'Friday')
    plt.plot(x, df_saturday['count'].values, label = 'Saturday')
    plt.plot(x, df_sunday['count'].values, label = 'Sunday')

    
    plt.legend(loc='upper right', prop={'size':15})
    plt.xlabel('Hours of the day', fontsize=15)
    plt.ylabel('Average number of rented bikes', fontsize=15)
    #plt.show()
    #plt.gcf().autofmt_xdate()
    plt.xticks(np.arange(0, len(x), 1))
   
    
    #plt.show()    
    fig1.savefig('rented bikes per week.png', format='png', dpi=300)
    
    #----
    y_count = df_ori['count']
    y_hour = df_ori['hour']
    y_day = df_ori['day_week']
    y_month = df_ori['month']
    
    fig1=plt.figure(figsize=(18,6))
    plt.plot(y_count[0:50], label='rented bikes', marker='o' , color='blue')
    plt.legend(loc='upper right', prop={'size':20})
    plt.xlabel('Time t', fontsize=20)
    plt.ylabel('Number of rented bikes', fontsize=20)
    plt.xticks(np.arange(0, 50, 2), fontsize=20, rotation=45)
    plt.yticks(fontsize=20)