import plotly.figure_factory as ff

z=[[1, 90, 30, 50, 1], [20, 1, 60, 80, 30], [30, 60, 1, 50, 20]]
x=['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday']
y=['Morning', 'Afternoon', 'Evening']

fig = ff.create_annotated_heatmap(z, x = x, y = y, colorscale = "Viridis")
fig.show()

import plotly.graph_objects as go
from functools import reduce
from itertools import product

z=[[1, 90, 30, 50, 1], [20, 1, 60, 80, 30], [30, 60, 1, 50, 20]]
x=['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday']
y=['Morning', 'Afternoon', 'Evening']

def get_anno_text(z_value):
    annotations=[]
    a, b = len(z_value), len(z_value[0])
    flat_z = reduce(lambda x,y: x+y, z_value) # z_value.flat if you deal with numpy
    coords = product(range(a), range(b))
    for pos, elem in zip(coords, flat_z):
        annotations.append({'font': {'color': '#FFFFFF'},
                    'showarrow': False,
                    'text': str(elem),
                    'x': pos[1],
                    'y': pos[0]})
    return annotations

fig = go.Figure(data=go.Heatmap(
                   z=z,
                   x=x,
                   y=y,
                   hoverongaps = True))

fig.update_layout(annotations = get_anno_text(z))
fig.show()

import plotly.graph_objects as go

z = df.groupby(['Q4', 'Q1']).size().unstack().fillna(0).astype('int16')
# convert to correlation matrix
z2 = z.apply(lambda x:x/x.sum(), axis = 1)

x = z2.columns
y = z2.index

fig = go.Figure(data=go.Heatmap(
                   z=z2.to_numpy(),  #dataframe을 넘파이(배열)로 바꿔줌: List형태
                   x=x,
                   y=y,
                   type="heatmap",
                   colorscale = "Viridis",
                   hoverongaps = False))

fig.update_layout(
    title='Degree ~ Gender',
    xaxis_nticks=36)

fig.show()

# 연도별 나이 
df21Age_Ea = df21_Ea.loc[:,['Q3','Q1']].reset_index().rename(columns={'Q3':'East_Asia', 'Q1':'2021'}).fillna('etc')
df20Age_Ea = df20_Ea.loc[:,['Q3','Q1']].reset_index().rename(columns={'Q3':'East_Asia', 'Q1':'2020'}).fillna('etc')
df19Age_Ea = df19_Ea.loc[:,['Q3','Q1']].reset_index().rename(columns={'Q3':'East_Asia', 'Q1':'2019'}).fillna('etc')
df18Age_Ea = df18_Ea.loc[:,['Q3','Q2']].reset_index().rename(columns={'Q3':'East_Asia', 'Q2':'2018'}).fillna('etc')
df17Age_Ea = df17_Ea.loc[:,['Country','Age']].reset_index().rename(columns={'Country':'East_Asia', 'Age':'2017'}).fillna('etc')


#data frame 정리

dfAge21_percent =df21Age_Ea.groupby(['East_Asia','2021']).size().reset_index().rename(columns = {0:"Count"})
dfAge21_percent['percent'] =((dfAge21_percent['Count'] / len(df21Age_Ea))*100).round(2)
dfAge21_percent['percent_str'] =((dfAge21_percent['Count'] / len(df21Age_Ea))*100).round(2).astype(str) + '%'
# dfAge21['percent'] = ((media['Count'] / len(df))*100).round(2).astype(str) + '%'
# dfAge_percent21=dfAge21.value_counts('East_Asia',normalize=True).mul(100).round(1).astype(str)
dfAge21_percent


# 나라별 연령대 비율 in East Asia
fig = go.Figure()
for country, group  in dfAge21_percent.groupby('East_Asia'):
   fig.add_trace(go.Bar(
       x = group['2021'], y = group['percent'], name = country, text=group['percent_str']
   ))
fig.update_layout(barmode="stack",
                 plot_bgcolor = "white",
                 title='2021_ 연령별 나라 비율 in East Asia')
fig.show()

# 연령별 나라 비율 in East Asia // 여기 name 어떻게 넣는거야 !!!! 
fig = go.Figure()

for country, group  in dfAge21_percent.groupby('2021'):
   fig.add_trace(go.Bar(
       x = group['East_Asia'], y =group['percent'], text=dfAge21_percent['percent_str']
   ))
fig.update_layout(barmode="stack",
                 plot_bgcolor = "white",
                 title='2021_ 나라별 연령 비율 in East Asia')
fig.show()

df21_Ko = df21[df21['Q3'] == 'South Korea']
df21_Wo = df21[~(df21['Q3'] == 'South Korea')]
df21['region']=["Korea" if x == 'South Korea' else "World" for x in df21['Q3']]
df21['region'].value_counts()

df20_Ko = df20[df20['Q3'] == 'South Korea']
df20_Wo = df20[~(df20['Q3'] == 'South Korea')]
df20['region']=["Korea" if x == 'South Korea' else "World" for x in df20['Q3']]
df20['region'].value_counts()

df19_Ko = df19[df19['Q3'] == 'South Korea']
df19_Wo = df19[~(df19['Q3'] == 'South Korea')]
df19['region']=["Korea" if x == 'South Korea' else "World" for x in df19['Q3']]
df19['region'].value_counts()

df18_Ko = df18[df18['Q3'] == 'South Korea']
df18_Wo = df18[~(df18['Q3'] == 'South Korea')]
df18['region']=["Korea" if x == 'South Korea' else "World" for x in df18['Q3']]
df18['region'].value_counts()

df17_Ko = df17[df17['Country'] == 'South Korea']
df17_Wo = df17[~(df17['Country'] == 'South Korea')]
df17['region']=["Korea" if x == 'South Korea' else "World" for x in df17['Country']]
df17['region'].value_counts()

## East Asia에는 대한민국, 일본, 중국, 타이완, 몽골, 북조선 총 6개의 국가가 속해 있다. 
## 알 수 없지만, 18년도엔 타이완이 없다. 

EastAsia17 = ['China',"People 's Republic of China", 'Taiwan', 'South Korea', 'Japan']
EastAsia18= ['China', 'South Korea', 'Japan', 'Republic of Korea'] 
EastAsia19 = ['China','Taiwan', 'South Korea', 'Japan', 'Republic of Korea']
EastAsia20 = ['China','Taiwan', 'South Korea','Republic of Korea', 'Japan']
EastAsia21 = ['China','Taiwan', 'South Korea', 'Japan']


EastAsia = ['Republic of Korea','China','Taiwan', 'South Korea', 'Japan', "People 's Republic of China" ]

df21_Ea = df21[df21['Q3'].isin(EastAsia)]
df21_Wo = df21[~df21['Q3'].isin(EastAsia )]
df21['region']=["EastAsia" if x in EastAsia else "World" for x in df21['Q3']]

df20_Ea = df20[df20['Q3'].isin(EastAsia)]
df20_Wo = df20[~df20['Q3'].isin(EastAsia )]
df20['region']=["EastAsia" if x in EastAsia else "World" for x in df20['Q3']]

df19_Ea = df19[df19['Q3'].isin(EastAsia)]
df19_Wo = df19[~df19['Q3'].isin(EastAsia )]
df19['region']=["EastAsia" if x in EastAsia else "World" for x in df19['Q3']]

df18_Ea = df18[df18['Q3'].isin(EastAsia)]
df18_Wo = df18[~df18['Q3'].isin(EastAsia )]
df18['region']=["EastAsia" if x in EastAsia else "World" for x in df18['Q3']]

df17_Ea = df17[df17['Country'].isin(EastAsia)]
df17_Wo = df17[~df17['Country'].isin(EastAsia )]
df17['region']=["EastAsia" if x in EastAsia else "World" for x in df17['Country']]


#df21['region'].to_frame().value_counts().to_frame().rename(columns={'region': '21y', '' : 'count'})

# % 계산을 위해 len() 을 통해 data 생성.

Ea21 = len(df21_Ea)
Wo21 = len(df21) - len(df21_Ea)

Ea20 = len(df20_Ea)
Wo20 = len(df20) - len(df20_Ea)

Ea19 = len(df19_Ea)
Wo19 = len(df19) - len(df19_Ea)

Ea18 = len(df18_Ea)
Wo18 = len(df18) -  len(df18_Ea)

Ea17 = len(df17_Ea)
Wo17 = len(df17) - len(df17_Ea)


def percent (a, b):
    result =a/(a+b)*100
    return result

def percentR (b, a):
    result =a/(a+b)*100
    return result

country = ['East Asia', 'Rest of the World']

years = ['2017', '2018', '2019', '2020', '2021']

fig = go.Figure(data=[
    go.Bar(name='Rest of the World', x=years, y=[percentR(Ea17, Wo17), percentR(Ea18, Wo18), percentR(Ea19, Wo19), 
                                                 percentR(Ea20, Wo20), percentR(Ea21, Wo21)]),
    go.Bar(name='East Asia', x=years, y=[percent(Ea17, Wo17), percent(Ea18, Wo18), percent(Ea19, Wo19), 
                                                 percent(Ea20, Wo20), percent(Ea21, Wo21)])
])

fig.update_layout(barmode='stack')
fig.show()

import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pylab as plt

import plotly.io as pio
import plotly.express as px
import plotly.graph_objects as go
import plotly.figure_factory as ff
from plotly.subplots import make_subplots
from plotly.offline import init_notebook_mode, iplot
init_notebook_mode(connected=True)
pio.templates.default = "none"
# import plotly.offline as py
# py.offline.init_notebook_mode()

import os
for dirname, _, filenames in os.walk('/kaggle/input'):
    for filename in filenames:
        print(os.path.join(dirname, filename))

import warnings
warnings.filterwarnings("ignore")

df17= pd.read_csv("/kaggle/input/kaggle-survey-2017/multipleChoiceResponses.csv", encoding="ISO-8859-1")
df18= pd.read_csv("/kaggle/input/kaggle-survey-2018/multipleChoiceResponses.csv", )
df19= pd.read_csv("/kaggle/input/kaggle-survey-2019/multiple_choice_responses.csv", )
df20= pd.read_csv("/kaggle/input/kaggle-survey-2020/kaggle_survey_2020_responses.csv", )
df21= pd.read_csv("/kaggle/input/kaggle-survey-2021/kaggle_survey_2021_responses.csv", )

df21 = df21.iloc[1:, :]
#df21.value_counts()
#df21.count

df21.head()

country = (
    df21['Q3']
    .value_counts()
    .to_frame()
    .reset_index()
    .rename(columns={'index':'country', 'Q3':'Count'})
    .sort_values(by=['country'], ascending=False)
     )  

Wage = (
    df21['Q25']
    .value_counts()
    .to_frame()
    .reset_index()
    .rename(columns={'index':'Wage', 'Q25':'Count'})
    .sort_values(by=['Wage'], ascending=True)
     )  

Wage

# India, Russia, China
temp = df21[['Q3','Q25']]

print(temp['Q3'].unique())

C_Ls = temp['Q3'].unique() #나라이름 뽑아주기

tempR = temp[temp['Q3'] == str(C_Ls[0])].value_counts()


# 이 경우에는 temp가 가공 할 수 없는 templet이기 때문에 plot을 그릴 수가 없다. 

#선생님
temp= df21[['Q3','Q25']]
temp = temp.groupby(['Q3','Q25'])
        .size()
        .reset_index()
        .rename(columns = {0:"Count"})

C_Ls = temp['Q3'].unique()

for i in range(1,len(df21['Q3'])):
    temp2 = temp[temp['Q3'] == str(C_Ls[i])]
    fig = px.bar(temp2, x='Q25', y='Count',title=C_Ls[i])
    fig.show()

def plotly_hBar(df, q, title, height=400,l=250,r=50,b=50,t=100,):   
    fig = px.histogram(df.iloc[1:], 
                       y=q,
                       orientation='h',
                       width=700,
                       height=height,
                       histnorm='percent',
                       color='region',
                       color_discrete_map={
                           "Africa": "gold", "World": "salmon"
                       },
                       opacity=0.6
                       )

    fig.update_layout(title=title, 
                      font_family="San Serif",
                      bargap=0.2,
                      barmode='group',
                      titlefont={'size': 28},
                      paper_bgcolor='#F5F5F5',
                      plot_bgcolor='#F5F5F5',                      
                      legend=dict(
                      orientation="v", 
                          y=1, 
                          yanchor="top", 
                          x=1.250, 
                          xanchor="right",)                 
                      ).update_yaxes(categoryorder='total ascending')
    
    fig.update_traces(marker_line_color='black',
                  marker_line_width=1.5)
    fig.update_layout(yaxis_title=None,yaxis_linewidth=2.5,
    autosize=False,
    margin=dict(
        l=l,
        r=r,
        b=b,
        t=t,
    ),
    )
    fig.update_xaxes(showgrid=False)
    fig.update_yaxes(showgrid=False)
    fig.show()   
    

africa17 = ['Nigeria','Kenya', 'South Africa', 'Egypt']
africa18 = ['Nigeria','Kenya', 'South Africa', 'Egypt', 'Tunisia', 'Morocco'] 
africa19 = ['Nigeria','Kenya', 'South Africa', 'Egypt', 'Tunisia', 'Morocco', 'Algeria']
africa20 = ['Nigeria','Kenya', 'South Africa', 'Egypt', 'Tunisia', 'Morocco', 'Ghana']
africa21 = ['Nigeria','Kenya', 'South Africa', 'Egypt', 'Tunisia', 'Morocco', 'Algeria', 'Ghana', 'Uganda', 'Ethiopia']

africa = ['Nigeria', 'Egypt', 'South Africa', 'Algeria', 'Tunisia', 'Morocco', 'Kenya', 'Uganda', 'Ghana', 'Ethiopia']

df21_africa = df21[df21['Q3'].isin(africa)]
df21_world = df21[~df21['Q3'].isin(africa )]
df21['region']=["Africa" if x in africa else "World" for x in df21['Q3']]

df20_africa = df20[df20['Q3'].isin(africa)]
df20_world = df20[~df20['Q3'].isin(africa )]
df20['region']=["Africa" if x in africa else "World" for x in df20['Q3']]

df19_africa = df19[df19['Q3'].isin(africa)]
df19_world = df19[~df19['Q3'].isin(africa)]
df19['region']=["Africa" if x in africa else "World" for x in df19['Q3']]

df18_africa = df18[df18['Q3'].isin(africa)]
df18_world = df18[~df18['Q3'].isin(africa)]
df18['region']=["Africa" if x in africa else "World" for x in df18['Q3']]

df17_africa = df17[df17['Country'].isin(africa)]
df17_world = df17[~df17['Country'].isin(africa )]
df17['region']=["Africa" if x in africa else "World" for x in df17['Country']]   

df21['region']=["Africa" if x in africa else "World" for x in df21['Q3']]