#!/usr/bin/env python
# coding: utf-8

# <a id="home"></a>

# # Székely(földi)ek pályája

# Ez a munkafüzet az [I. SZÉKELYDATA adatvizualizációs pályázat](http://csaladenes.egologo.ro/?p=1129) keretén belül publikált adatsort dolgozza fel, tisztítja, és készíti elő későbbi vizualizáláshoz, mely [ezen] (https://lorinczszabolcs.github.io/eletpalya/) az oldalon megtekinthető. 
# 
# Több lépésben végezzük az elemzést:
# 
# 1. [Adatok feltérképezése és tisztítása](#1)
#     1. [Születési helyek feldolgozása](#11)
#     2. [Tanulmányi területek feldolgozása](#12)
#     3. [Szakterületek feldolgozása](#13)
#     4. [Lakhelyek feldolgozása](#14)
# 2. [Korreláció vizsgálata](#2)
#     1. [születési hely - tanulmányok között](#21)
#     2. [születési hely - szakterületek között](#22)
#     3. [lakhely - tanulmányok között](#23)
#     4. [lakhely - szakterületek között](#24)
#     5. [tanulmányok - szakterületek között](#25)
# 3. [Adatok előkészítése vizualizációhoz](#3)
# </span>

# ##  0. Kezdeti beállítások, szükséges könyvtárak importálása
# 
# [Tetejére! &#8673;](#home)

# In[143]:


import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
import json

# unidecode - ékezetek eltűntetése
import unidecode

# stílus a megjelenítéshez
plt.style.use('ggplot')


# In[144]:


# varosok lakossaga
lakossag = {'Kolozsvár': 324576, 'Kovászna': 10114,
            'Marosvásárhely': 134290, 'Csíkszereda' : 38966,
            'Szentgyörgy': 56006, 'Gyergyó': 18377,
            'Kézdivásárhely': 18491, 'Székelyudvarhely': 32935,
            'Székelykeresztúr': 9491, 'Szászrégen': 33281,
            'Maroshévíz': 13929, 'Segesvár': 28102,
            'Barót': 8672, 'Szentegyházasfalu': 6898,
            'Balánbánya': 6115, 'Szováta': 10385,
            'Parajd': 3448}


# ## 1. Adatok feltérképezése és tisztítása <a id='1'></a>
# [Tetejére! &#8673;](#home)

# In[145]:


filename = '../data/data.json'
df = pd.read_json(filename)
df.info()


# Úgy tűnik elég **sok hiányzó adattal** van dolgunk. Ezeket majd a későbbiekben ki kell szűrjük.

# ### Tanulmányi terület, szakterület és lakhely feltérképezése
# [Tetejére! &#8673;](#home)

# In[146]:


fig = plt.figure(figsize=(5,5));
stud_class=df['studwhat'].value_counts()
ax = stud_class[stud_class > 10].plot.pie(autopct='%.1f')
ax.axis('equal');
titledict = {'fontsize': 15, 'fontweight' : 100, 'verticalalignment': 'baseline', 'horizontalalignment': 'center'};
ax.set_title('Tanulmányi területek (>10 ember)', fontdict=titledict, y = 1.08);
ax.set_ylabel('');
print('Tanulmányi területek száma: ', df['studwhat'].value_counts().size)


# In[147]:


fig = plt.figure(figsize=(5,5));
work_class=df['workwhat'].value_counts()
ax = work_class[work_class > 10].plot.pie(autopct='%.1f')
ax.axis('equal');
titledict = {'fontsize': 15, 'fontweight' : 100, 'verticalalignment': 'baseline', 'horizontalalignment': 'center'};
ax.set_title('Szakterületek (>10 ember)', fontdict=titledict, y = 1.08);
ax.set_ylabel('');
print('Szakterületek száma: ', df['workwhat'].value_counts().size)


# In[148]:


fig = plt.figure(figsize=(5,5));
live_class=df['livesin'].value_counts()
ax = live_class[live_class > 15].plot.pie(autopct='%.1f')
ax.axis('equal');
titledict = {'fontsize': 15, 'fontweight' : 100, 'verticalalignment': 'baseline', 'horizontalalignment': 'center'};
ax.set_title('Lakhelyek (>15 ember)', fontdict=titledict, y = 1.08);
ax.set_ylabel('');
print('Lakhelyek száma: ', df['livesin'].value_counts().size)


# Átnevezzük a születési helyeket, majd **tisztítani és csoportosítani** kell a tanulmányi területeket, munkahelyeket, és lakhelyeket, mivel több nyelven, helyesírási hibával, vagy anélkül is szerepelnek adatok.

# ### 1.A. Születési helyek feldolgozása <a id="11"></a>
# [Tetejére! &#8673;](#home)
# 
# - Születési helyek átnevezése

# In[149]:


def birth(b):
    """
    Rename birth
    """
    
    if b == 'balan' : return 'Balánbánya'
    elif b == 'barot' : return 'Barót'
    elif b == 'csikszereda' : return 'Csíkszereda'
    elif b =='gyergyo' : return 'Gyergyó'
    elif b == 'keresztur' : return 'Székelykeresztúr'
    elif b == 'kezdi' : return 'Kézdivásárhely'
    elif b == 'kolozsvar' : return 'Kolozsvár'
    elif b == 'kovaszna' : return 'Kovászna'
    elif b == 'parajd' : return 'Parajd'
    elif b == 'regen' : return 'Szászrégen'
    elif b == 'segesvar' : return 'Segesvár'
    elif b == 'szentgyorgy' : return 'Szentgyörgy'
    elif b == 'szovata' : return 'Szováta'
    elif b == 'toplita' : return 'Maroshévíz'
    elif b == 'udvarhely' : return 'Székelyudvarhely'
    elif b == 'vasarhely' : return 'Marosvásárhely'
    elif b == 'vlahica' : return 'Szentegyházasfalu'
    else : return b


# In[150]:


df[['birth']] = df[['birth']].applymap(lambda x : birth(x))
print(df['birth'].value_counts())


# ### 1.B. Tanulmányi területek feldolgozása <a id="12"></a>
# [Tetejére! &#8673;](#home)
# 
# - ékezetek törlése
# - helyesírási hibák kiszűrése
# - nyelvtől független csoportosítás

# In[151]:


df[['studwhat']] = df[['studwhat']].applymap(lambda x : str.lower(unidecode.unidecode(x)) if pd.notna(x) else np.nan)


# In[152]:


def stud(s):
    """
    Categorises input field of study 's' into main categories.
    """
    
    # Mérnöki/Technológia
    if 'engineer' in s \
        or 'inginer' in s \
        or 'tcm' in s \
        or 'mernok' in s \
        or 'archite' in s \
        or 'a.c.' in s \
        or 'urbanism' in s \
        or 'industr' in s \
        or 'constructii' in s \
        or 'ccia' in s \
        or 'ing.' in s \
        or 'cepa' in s \
        or 'robot' in s \
        or 'instalatii' in s \
        or 'securitatea rutiera' in s \
        or 'upb' in s \
        or 'utc-n' in s \
        or 'automotive' in s \
        or 'mecha' in s \
        or 'cad' in s \
        or 'electro' in s \
        or 'hardware' in s \
        or 'mechanics' in s \
        or 'autovehicule rutiere' in s \
        or 'mecanica' in s \
        or 'mehanika' in s: return u'Mérnöki/Technológia'
    # Matematika/Informatika
    elif 'computer' in  s \
        or 'software' in s \
        or 'intelligent systems' in s \
        or 'szamitas' in s \
        or 'calculatoare' in s \
        or 'info' in s \
        or 'cig' in s \
        or 'database' in s \
        or 'mate' in s \
        or 'metematika' in s \
        or 'szoft' in s \
        or 'it' in s \
        or 'math' in s : return u'Matematika/Informatika'
    # Természettudományok
    elif 'biol' in  s \
        or 'geo' in s \
        or 'astronomy' in s \
        or 'okologia' in s \
        or 'kornyezettudomany' in s \
        or 'foldrajz' in s \
        or 'meteorology' in s \
        or 'meteorologia' in s \
        or 'chemistry' in s \
        or 'fizika' in s \
        or 'environmental science' in s \
        or 'stiinte ale naturi' in s \
        or 'topography' in s \
        or 'physics' in s \
        or 'chimie' in s \
        or 'kemia' in s \
        or 'natural science' in s : return u'Természettudományok'
    # Orvosi/Gyógyszerészet/Egészség
    elif 'medic' in  s \
        or 'nurs' in s \
        or 'pharmacy' in s \
        or 'etszk' in s \
        or 'pharmac' in s \
        or 'beteg' in s \
        or 'kineto' in s \
        or 'balneo' in s \
        or 'apolo' in  s\
        or 'egeszseg' in s \
        or 'dentistry' in s \
        or 'fogtechnika' in s \
        or 'health' in s \
        or 'fogasz' in s \
        or 'dent' in s \
        or 'aok' in s \
        or 'nutrition & dietetics' in s \
        or 'sykepleie bachelor' in s \
        or 'orvos' in s \
        or 'gyogy' in s \
        or 'kinetoterapie' in s : return u'Egészségügy'
    # Pénzügy/Management/Üzlet
    elif 'management' in  s \
        or 'marketing' in s \
        or 'finan' in s \
        or 'bussiness' in s \
        or 'kereskedelem' in s \
        or 'manag' in s \
        or 'konyvel' in s \
        or 'human resource' in s \
        or 'hr' in s \
        or 'business' in s \
        or 'contabilitate' in s \
        or 'administratie' in s \
        or 'fsega' in s \
        or 'acounting' in s \
        or 'bank' in s \
        or 'account' in s \
        or 'betriebswirtschaft' in s \
        or 'afaceri' in s : return u'Pénzügy/Management/Üzlet'
     # Társadalomtudományok
    elif 'sociology' in  s \
        or 'econom' in s \
        or 'social science' in s \
        or 'psycho' in s \
        or 'pszicho' in s \
        or 'psziho' in s \
        or 'psiho' in s \
        or 'mental' in s \
        or 'gazdalkodas' in s \
        or 'kozgazdasz' in s \
        or 'gazdasag' in s \
        or 'tarsadalmi' in s \
        or 'law' in  s\
        or 'cultural' in s \
        or 'juridice' in s \
        or 'tortenelem' in s \
        or 'jog' in s \
        or 'drept' in s \
        or 'ecology' in s \
        or 'touris' in s \
        or 'social' in s \
        or 'turis' in s \
        or 'turiz' in s \
        or 'international relations' in  s \
        or 'nemzetkozi kapcsolatok' in s \
        or 'african studies' in s \
        or 'politic' in s \
        or 'kozigazgatas' in s \
        or 'archaeology' in s \
        or 'public administration' in s \
        or 'r.i.s.e' in s \
        or 'studii europene' in s \
        or 'nemzetkozi tanulmanyok' in s \
        or 'european studies' in s \
        or 'polito' in s \
        or 'public policy' in s \
        or 'socio' in s \
        or 'szocio' in s : return u'Társadalomtudományok'
    # Oktatás
    elif 'peda' in  s \
        or 'teaching' in s \
        or 'tanar' in s \
        or 'andragog' in s \
        or 'alternativ modszerek az ovodaban es az elemi osztalyokban' in s \
        or 'neveles' in s \
        or 'd.p.p.d' in s \
        or 'stiinte ale educatiei' in s \
        or 'educatie fizica' in s \
        or 'ed.fiz' in s \
        or 'bodybuilding' in s \
        or 'physical' in s \
        or 'volei' in s \
        or 'testnevel' in s \
        or 'education' in s : return u'Oktatás'
    # Művészetek
    elif 'photography' in  s \
        or 'photo' in s \
        or 'acting' in s \
        or 'operator' in s \
        or 'saxophone' in s \
        or 'graphics' in s \
        or 'art' in s \
        or 'canto popular' in s \
        or 'actor' in s \
        or 'muveszet' in s \
        or 'choreography' in s \
        or 'painting' in s \
        or 'music' in s \
        or 'desiner' in s \
        or 'szobraszat' in s \
        or 'puppery' in s \
        or 'muzica' in s \
        or 'piano' in s \
        or 'violin' in s \
        or 'sing' in s \
        or 'zene' in s \
        or 'drama' in s \
        or 'bass' in s \
        or 'szinesz' in s \
        or 'percussion' in s \
        or 'puppetry' in s \
        or 'dance' in s \
        or 'sculpture' in s \
        or 'szinmuvesz' in s \
        or 'kameraman' in s \
        or 'enek' in s : return u'Művészetek'
    # Média/Kommunikáció
    elif 'journalism' in  s\
        or 'ujsag' in s \
        or 'design' in s \
        or 'advertising' in s \
        or 'com' in s \
        or 'film' in s \
        or 'public relations' in s \
        or 'pr' in s \
        or 'newspaper' in s \
        or 'media' in s \
        or 'komm' in s : return u'Média/Kommunikáció'
    # Humán tárgyak
    elif 'roman' in s \
        or 'engleza' in s \
        or 'english' in s \
        or 'german' in s \
        or 'literature' in s \
        or 'translation' in s \
        or 'angol' in s \
        or 'language' in s \
        or 'limbi' in s \
        or 'hungarian' in s \
        or 'linguistics' in s \
        or 'japanese' in s \
        or 'bolcsesz' in s \
        or 'magyar' in s \
        or 'aesthetics' in s \
        or 'deutsch' in s \
        or 'french' in s \
        or 'history'in s \
        or 'philosophy' in s \
        or 'bolcseszkar' in s \
        or 'filozofia' in s \
        or 'katolikus' in s \
        or 'phylosophy' in s \
        or 'philology' in s \
        or 'filologia' in s \
        or 'filozofie' in s \
        or 'religi' in s \
        or 'missions & evangelism' in s \
        or 'theology' in s \
        or 'vallas' in s : return u'Humán tárgyak'
    # Szakma/Alkalmazott tudományok
    elif 'asztalos' in s \
        or 'hivatasos' in s \
        or 'ospatar' in s \
        or 'auto electrician' in s \
        or 'pincer' in s \
        or 'agriculture' in s \
        or 'vezetes es szervezes' in s \
        or 'szakacs' in s \
        or 'foci' in s \
        or 'kozmetikus' in s \
        or 'hairdressing' in s \
        or 'coafura' in s \
        or 'kozmetika' in s \
        or 'optometry' in s \
        or 'fodrasz' in s \
        or 'sport' in s \
        or 'cosmetology' in s \
        or 'lacatus' in s \
        or 'lacatuserie' in s \
        or 'lakatos' in s \
        or 'gepesz' in s \
        or 'autoszerelo' in s \
        or 'agro' in s \
        or 'horticulture' in s \
        or 'silvicultura' in s \
        or 'silviculture' in s \
        or 'faipar' in s \
        or 'optometria' in s : return u'Alkalmazott tudományok'
    else : return s


# In[153]:


fig = plt.figure(figsize=(5,5));
df[['studwhat']] = df[['studwhat']].applymap(lambda x : stud(x) if pd.notna(x) else np.nan)
stud_class=df['studwhat'].value_counts()
ax = stud_class[stud_class > 15].plot.pie(autopct='%.1f')
ax.axis('equal');
titledict = {'fontsize': 15, 'fontweight' : 100, 'verticalalignment': 'baseline', 'horizontalalignment': 'center'};
ax.set_title('Tanulmányi területek (>15 ember)', fontdict=titledict, y = 1.08);
ax.set_ylabel('');
print('Tanulmányi területek száma: ', stud_class[stud_class > 15].value_counts().size)


# ### 1.C. Szakterületek feldolgozása <a id="13"></a>
# [Tetejére! &#8673;](#home)
# - ékezetek törlése
# - helyesírási hibák kiszűrése
# - nyelvtől független csoportosítás

# In[154]:


df[['workwhat']] = df[['workwhat']].applymap(lambda x : str.lower(unidecode.unidecode(x)) if pd.notna(x) else np.nan)


# In[155]:


df['workwhat'].value_counts()


# In[156]:


def work(w):
    """
    Categorises input work 'w' into main categories.
    """
    # Programmer
    if 'software' in w \
        or 'developer' in w \
        or 'informa' in w \
        or 'asm' in w \
        or 'sysop' in w \
        or 'it compliance' in w \
        or 'programmer' in w \
        or 'web' in w : return u'Informatikus'
    elif 'tanitono' in w \
        or 'tanar' in w \
        or 'ovono' in w \
        or 'teacher' in w \
        or 'dirigente' in w \
        or 'pedagog' in w \
        or 'student' in w \
        or 'tanulo' in w \
        or 'invatatoare' in w \
        or 'tanito' in w \
        or 'sudent' in w \
        or 'elev' in w \
        or 'muvelodesszervezo' in w \
        or 'oktato' in w \
        or 'teaching' in w \
        or 'study' in w \
        or 'diak' in w \
        or 'ovoneni' in w \
        or 'profesor' in w \
        or 'lecturer' in w \
        or 'instructor' in w \
        or 'university' in w \
        or 'adjunktus' in w \
        or 'educatoare' in w \
        or 'researcher' in w \
        or 'professor' in w : return 'Oktatás'
    elif 'sofer' in w \
        or 'driver' in w \
        or 'curier' in w \
        or 'posta' in w \
        or 'gepkocsivezeto' in w \
        or 'raktar' in w \
        or w == 'vezeto' \
        or 'komissio' in w \
        or 'messenger' in w \
        or 'komisszio' in w \
        or 'comission' in w \
        or 'fahrer' in w \
        or 'transport' in w \
        or 'rider' in w \
        or 'sofor' in w : return 'Logisztika'
    elif 'volunteer' in w \
        or 'au pair' in w \
        or 'voluntar' in w \
        or 'social work' in w \
        or 'onkentes' in w : return 'Önkéntes'
    elif 'tancos' in w \
        or 'szinesz' in w \
        or 'enekes' in w \
        or 'drummer' in w \
        or 'artist' in w \
        or 'dj' in w \
        or 'music' in w \
        or 'guitar' in w \
        or 'muvesz' in w \
        or 'musician' in w \
        or 'bassist' in w \
        or 'zenesz' in w \
        or 'karmester' in w \
        or 'actress' in w \
        or 'animator' in w \
        or 'actor' in w \
        or 'dance' in w \
        or 'grafik' in w \
        or 'sing' in w \
        or 'actor' in w : return 'Művész'
    elif 'sales' in w \
        or 'consultant' in w \
        or 'uzlet' in w \
        or 'marketing' in w \
        or 'cashier' in w \
        or 'casier' in w \
        or 'ugynok' in w \
        or 'agent' in w \
        or 'broker' in w \
        or 'vanzari' in w \
        or 'vanzato' in w \
        or 'elado' in w : return 'Eladás'
    elif 'bartender' in w \
        or 'waitress' in w \
        or 'pultos' in w \
        or 'pincer' in w \
        or 'reception' in w \
        or 'kiszolgalo' in w \
        or 'ospatar' in w \
        or 'barista' in w \
        or 'bucatar' in w \
        or 'waiter' in w \
        or 'waiting staff' in w \
        or 'baros' in w \
        or 'cukrasz' in w \
        or 'sef' in w \
        or 'szakacs' in w \
        or 'chef' in w \
        or 'barman' in w \
        or 'pincer' in w : return 'Vendéglátás'
    elif 'asistent' in w \
        or 'orvos' in w \
        or 'aszisztens' in w \
        or 'gyogyszer' in w \
        or 'psiholog' in w \
        or 'health' in w \
        or 'apolo' in w \
        or 'veterinarian' in w \
        or 'kranken' in w \
        or 'assistenz' in w \
        or 'nurse' in w \
        or 'assziszt' in w \
        or 'orthodontist' in w \
        or 'dietitian' in w \
        or 'physician' in w \
        or 'dr.' in w \
        or 'rezidens' in w \
        or 'kineto' in w \
        or 'dentist' in w \
        or 'pszicho' in w \
        or 'rezident' in w \
        or 'doctor' in w \
        or 'pharma' in w \
        or 'optometrista' in w \
        or 'farmac' in w \
        or 'medic' in w : return 'Egészségügy'
    elif 'kozmetik' in w \
        or 'fodrasz' in w \
        or 'kormos' in w \
        or 'manichiur' in w \
        or 'coaf' in w \
        or 'hair' in w : return 'Kozmetika/Fodrász'
    elif 'train' in w \
        or 'torna' in w \
        or 'versenyzo' in w \
        or 'jucator' in w \
        or 'forward' in w \
        or 'vego' in w \
        or 'beallo' in w \
        or 'kozeppalyas' in w \
        or 'jatekos' in w \
        or 'coach' in w \
        or 'striker' in  w \
        or 'arbitru' in w \
        or 'midfielder' in w \
        or 'sport' in w \
        or 'captain' in w \
        or 'player' in w \
        or 'fotbal' in w \
        or 'atacant' in w \
        or 'csatar' in w \
        or 'goalkeeper' in w \
        or 'fitness' in w \
        or 'antrenor' in w : return 'Sport'
    elif 'secretary' in w \
        or 'alkalmazott' in w \
        or 'iroda' in w \
        or 'oordinator' in w \
        or 'economist' in w \
        or 'titkarno' in w \
        or 'coordonator' in w \
        or 'public relations' in w \
        or 'contabil' in w \
        or 'konyvtaros' in w \
        or 'kozgazdasz' in w \
        or 'referens' in w \
        or 'secretara' in w \
        or 'technical support' in w \
        or 'dispecer' in w \
        or 'titkar' in w \
        or 'inspector' in w \
        or 'referent' in w \
        or 'partner' in w \
        or 'employee' in w \
        or 'accountant' in w \
        or 'konyvelo' in w : return 'Irodai alkalmazott' 
    elif 'engineer' in w \
        or 'inginer' in w \
        or 'mernok' in w \
        or 'epitesz' in w \
        or 'constructor' in w \
        or 'arhitect' in w \
        or 'architect' in w \
        or 'arhchitect' in w \
        or 'autocad' in w : return 'Mérnök'
    elif 'polgarmester' in w \
        or 'president' in w \
        or 'tanacsos' in w \
        or 'orsvezeto' in w \
        or 'presedinte' in w : return 'Közalkalmazott'
    elif 'photo' in w \
        or 'media' in w \
        or 'designer' in w \
        or 'operator' in w \
        or 'foto' in w \
        or 'writer' in w \
        or 'model' in w \
        or 'musorvezeto' in w \
        or 'press' in w \
        or 'reporter' in w \
        or 'editor' in w \
        or 'szerkeszto' in w \
        or 'ujsagiro' in w \
        or 'producer' in w \
        or 'blogger' in w \
        or 'journalist' in w \
        or 'martketing' in w : return 'Média'
    elif 'lawyer' in w \
        or 'ugyved' in w \
        or 'juridic' in w \
        or 'avocat' in w : return 'Jogi'
    elif 'varrono' in w \
        or 'munkas' in w \
        or 'muncitor' in w \
        or 'electrician' in w \
        or 'villanyszerelo' in w \
        or 'trabajador' in w \
        or 'kazan' in w \
        or 'worker' in w \
        or 'mindenes' in w \
        or 'kasszas' in w \
        or 'gepkezelo' in w \
        or 'flight attendant' in w \
        or 'lucrator comercial' in w \
        or 'szerelo' in w \
        or 'baker' in w \
        or 'muncitoare' in w \
        or 'cabin crew' in w \
        or 'elarusito' in w \
        or 'technician' in w \
        or 'pilot' in w \
        or 'masszor' in w \
        or 'pastor' in w \
        or 'mechanic' in w \
        or 'firefighter' in w \
        or 'mecanic' in w \
        or 'hegeszto' in w \
        or 'fafarago' in w \
        or 'asztalos' in w : return 'Szakmunkás'
    elif 'boss' in w \
        or 'owner' in w \
        or 'manager' in w \
        or 'ugyvezeto' in w \
        or 'manager' in w \
        or 'admin' in w \
        or 'igazgato' in w \
        or 'vallalkozo' in w \
        or 'presedinte' in w \
        or 'masseur' in w \
        or 'chief' in w \
        or 'president' in w \
        or 'cegvezeto' in w \
        or 'leader' in w \
        or 'ceo' in w \
        or 'presedinte' in w \
        or 'supervisor' in w \
        or 'director' in w \
        or 'tulajdonos' in w \
        or 'menedzser' in w \
        or 'elnok' in w \
        or 'fonok' in w \
        or 'management' in w \
        or 'founder' in w : return 'Tulajdonos/Manager'
    elif 'pap' in w \
        or 'popa' in w \
        or 'lelkesz' in w \
        or 'priest' in w \
        or 'kantor' in w \
        or 'lelkipasztor' in w : return 'Egyház'
    else: return np.nan


# In[157]:


fig = plt.figure(figsize=(5,5));
df[['workwhat']] = df[['workwhat']].applymap(lambda x : work(x) if pd.notna(x) else np.nan)
work_class=df['workwhat'].value_counts()
ax = work_class[work_class > 15].plot.pie(autopct='%.1f')
ax.axis('equal');
titledict = {'fontsize': 15, 'fontweight' : 100, 'verticalalignment': 'baseline', 'horizontalalignment': 'center'};
ax.set_title('Szakterületek (>15 ember)', fontdict=titledict, y = 1.08);
ax.set_ylabel('');


# ### 1.D. Lakhelyek feldolgozása <a id="14"></a>
# [Tetejére! &#8673;](#home)
# - ékezetek törlése
# - helyesírási hibák kiszűrése
# - nyelvtől független csoportosítás

# In[158]:


df[['livesin']] = df[['livesin']].applymap(lambda x : str.lower(unidecode.unidecode(x)) if pd.notna(x) else np.nan)


# In[159]:


df['livesin'].value_counts()


# In[160]:


def lives(l) :
    if 'miercurea-ciuc' in l \
        or 'miercurea ciuc' in l \
        or 'csikszereda' in l : return 'Csíkszereda'
    elif 'szekelyudvarhely' in l \
        or 'odorheiul secuiesc' in l \
        or 'odorheiu secuiesc' in l: return 'Székelyudvarhely'
    elif 'gheorgheni' in l \
        or 'gyergyo' in l : return 'Gyergyó'
    elif 'targu-mures' in l \
        or 'tirgu-mures' in l \
        or 'marosvasarhely' in l : return 'Marosvásárhely'
    elif 'sepsiszentgyorgy' in l \
        or 'sfantu-gheorghe' in l \
        or 'sfantu gheorghe' in l : return 'Sepsiszentgyörgy'
    elif 'sighisoara' in l \
        or 'segesvar' in l : return 'Segesvár'
    elif 'cluj-napoca' in l \
        or 'kolozsvar' in l :return 'Kolozsvár'
    elif 'kezdivasarhely' in l \
        or 'targu secuiesc' in l : return 'Kézdivásárhely'
    elif 'oradea' in l \
        or 'nagyvarad' in l : return 'Nagyvárad'
    elif 'timisoara' in l \
        or 'temesvar' in l : return 'Temesvár'
    elif 'brasov' in l \
        or 'brasso' in l : return 'Brassó'
    elif 'bucharest' in l \
        or 'bukarest' in l \
        or 'bucuresti' in l : return 'Bukarest'
    elif 'budapest' in l : return 'Budapest'
    elif 'reghin' in l \
        or 'regen' in l : return 'Szászrégen'
    elif 'keresztur' in l \
        or 'cristuru' in l : return 'Székelykeresztúr'
    elif 'toplita' in l \
        or 'toplica' in l : return 'Toplița'
    elif 'szentegyhaza' in l \
        or 'vlahita' in l : return 'Szentegyházasfalu'
    elif 'london' in l : return 'London'
    elif 'sovata' in l \
        or 'szovata' in l : return 'Szováta'
    elif 'barot' in l \
        or 'baraolt' in l : return 'Barót'
    elif 'balan' in l : return 'Balánbánya'
    elif 'parajd' in l \
        or 'praid' in l : return 'Parajd'
    elif 'arad' in l : return 'Arad'
    elif 'vienna' in l \
        or 'becs' in l : return 'Bécs'   
    elif 'dublin' in l : return 'Dublin'
    elif 'tusnad' in l : return 'Tusnád'
    elif 'sibiu' in l \
        or 'szeben' in l : return 'Nagyszeben'
    elif 'gyulafehervar' in l \
        or 'alba iulia' in l : return 'Gyulafehérvár'
    elif 'iasi' in l \
        or 'jaszvasar' in l : return 'Jászvásár'
    elif 'munich' in l \
        or 'munchen' in l : return 'München'
    elif 'roma' in l \
        or 'rome' in l : return 'Róma'
    elif 'new york' in l : return 'New York'
    elif 'turia' in l \
        or 'torja' in l : return 'Torja'
    elif 'berlin' in l : return 'Berlin'
    elif 'debrecen' in l : return 'Debrecen'
    elif 'sopron' in l : return 'Sopron'
    elif 'szeged' in l : return 'Szeged'
    elif 'ozsdola' in l \
        or 'ojdula' in l : return 'Ozsdola'
    elif 'dumbraveni' in l : return 'Dumbrăveni'
    elif 'vecses' in l : return 'Vecsés'
    elif 'kaposvar' in l : return 'Kaposvár'
    elif 'mosonmagyarovar' in l : return 'Mosonmagyaróvár'
    elif 'eger' in l : return 'Eger'
    elif 'gelence' in l \
        or 'ghelinta' in l : return 'Gelence'
    elif 'borsec' in l \
        or 'borszek' in l : return 'Borszék'
    elif 'galati' in l \
        or 'galac' in l : return 'Galați'
    elif 'deva' in l : return 'Déva'
    elif 'lajosmizse' in l : return 'Lajosmizse'
    elif 'erd' in l : return 'Érd'
    elif 'toronto' in l : return 'Toronto'
    elif 'veszprem' in l : return 'Veszprém'
    elif 'cologne' in l \
        or 'koln' in l : return 'Köln'
    elif 'pusztavam' in l : return 'Pusztavám'
    elif 'intorsura buzaului' in l \
        or 'bodzafordulo' in l : return 'Bodzaforduló'
    elif 'bistrita' in l \
        or 'beszterce' in l : return 'Beszterce'
    elif 'tarnaveni' in l \
        or 'dicsoszentmarton' in l : return 'Târnăveni'
    elif 'lazarea' in l \
        or 'szarhegy' in l : return 'Szárhegy'
    elif 'kolding' in l : return 'Kolding'
    elif 'passau' in l : return 'Passau'
    elif 'ingolstadt' in l : return 'Ingolstadt'
    elif l == 'covasna' \
        or l == 'kovaszna' : return 'Kovászna'
    elif 'satu mare' in l \
        or 'szatmar' in l : return 'Szatmár'
    elif 'atlanta' in l : return 'Atlanta'
    elif 'pecs' in l : return 'Pécs'
    elif 'korinth' in l : return 'Korinth'
    elif 'kerepes' in l : return 'Kerepes'
    elif 'nyiregyhaza' in l : return 'Nyíregyháza'
    elif 'gidofalva' in l \
        or 'ghidfalau' in l : return 'Gidófalva'
    elif 'piatra neamt' in l : return 'Piatra Neamț'
    else: return l


# In[161]:


fig = plt.figure(figsize=(5,5));
df[['livesin']] = df[['livesin']].applymap(lambda x : lives(x) if pd.notna(x) else np.nan)
live_class=df['livesin'].value_counts()
ax = live_class[live_class > 100].plot.pie(autopct='%.1f')
ax.axis('equal');
titledict = {'fontsize': 15, 'fontweight' : 100, 'verticalalignment': 'baseline', 'horizontalalignment': 'center'};
ax.set_title('Lakhelyek (>100 ember)', fontdict=titledict, y = 1.08);
ax.set_ylabel('');
print('Lakhelyek száma: ', live_class[live_class > 100].size)


# ## 2. Korreláció vizsgálata <a id="2"></a>
# [Tetejére! &#8673;](#home)

# ### 2.a. Korreláció vizsgálata születési hely és tanulmányok között <a id="21"></a>
# [Tetejére! &#8673;](#home)

# Előbb kiválasztjuk a megfelelő oszlopokat, majd töröljük az összes olyan mérést, mely nincs kitöltve(NaN), valamint töröljük az ékezeteket és kisbetűsre alakítjuk a DataFrame tartalmát, hogy könnyebb legyen később a feldolgozás.

# In[162]:


df_birth_stud=df.loc[:, ['birth', 'studwhat']].copy()
df_birth_stud.dropna(inplace=True)
df_birth_stud = df_birth_stud.groupby('studwhat').filter(lambda x : len (x) > 15)


# Adatok mennyisége százalékban:

# In[163]:


[[df_birth_stud.iloc[:, 0].value_counts().index[i], 100*df_birth_stud.iloc[:, 0].value_counts()[i]/lakossag[df_birth_stud.iloc[:, 0].value_counts().index[i]], df_birth_stud.iloc[:, 0].value_counts()[i]] for i in range(df_birth_stud.iloc[:, 0].value_counts().size)]


# In[164]:


heatmap = pd.crosstab(df_birth_stud.birth, df_birth_stud.studwhat)


# A lenti hőtérképről könnyen leolvasható, hogy egyes városokban született emberek esetén mely tanulmányi területek a legnépszerűbbek.
# 
# **További elemzés a [weboldalon](https://lorinczszabolcs.github.io/eletpalya/).**

# In[165]:


fig, ax = plt.subplots(figsize=(15,10))
ax = sns.heatmap(heatmap, annot=True, cbar=False, fmt='g', annot_kws={"size": 14})
ax.set_title('Születési hely - Tanulmányok', fontsize=24, y = 1.04);
ax.set_xlabel('Tanulmányok', fontsize=18);
ax.set_ylabel('Születési hely', fontsize=18);
plt.setp(ax.get_xticklabels(), rotation='vertical', fontsize=14);
plt.setp(ax.get_yticklabels(), rotation='horizontal', fontsize=14);


# ### 2.b. Korreláció vizsgálata születési hely és szakterületek között <a id="22"></a>
# [Tetejére! &#8673;](#home)

# In[166]:


df_birth_work=df.loc[:, ['birth', 'workwhat']].copy()
df_birth_work.dropna(inplace=True)
df_birth_work = df_birth_work.groupby('workwhat').filter(lambda x : len (x) >= 10)


# Adatok mennyisége százalékban a népességhez képest:

# In[167]:


[[df_birth_work.iloc[:, 0].value_counts().index[i], 100*df_birth_work.iloc[:, 0].value_counts()[i]/lakossag[df_birth_work.iloc[:, 0].value_counts().index[i]]] for i in range(df_birth_work.iloc[:, 0].value_counts().size)]


# In[99]:


heatmap2 = pd.crosstab(df_birth_work.birth, df_birth_work.workwhat)
heatmap2_norm_col=(heatmap2-heatmap2.mean())/heatmap2.std()


# In[100]:


fig, ax = plt.subplots(figsize=(15,10))
ax = sns.heatmap(heatmap2, annot=True, cbar=False, fmt='g', annot_kws={"size": 14})
ax.set_title('Születési hely - szakterület', fontsize = 24, y=1.04)
ax.set_xlabel('Szakterület', fontsize=18);
ax.set_ylabel('Születési hely', fontsize=18);
plt.setp(ax.get_xticklabels(), rotation='vertical', fontsize=14);
plt.setp(ax.get_yticklabels(), rotation='horizontal', fontsize=14);


# Normalizálva oszlop szerint, mert a manager miatt a többi nem szembetűnő.

# In[101]:


fig, ax = plt.subplots(figsize=(15,10))
ax = sns.heatmap(heatmap2_norm_col, annot=True, cbar=False, annot_kws={"size": 14})
ax.set_title('Születési hely - szakterület (oszlop szerint normalizálva)', fontsize = 24, y = 1.04)
ax.set_xlabel('Szakterület', fontsize=18);
ax.set_ylabel('Születési hely', fontsize=18);
plt.setp(ax.get_xticklabels(), rotation='vertical', fontsize=14);
plt.setp(ax.get_yticklabels(), rotation='horizontal', fontsize=14);


# ### 2.c. Korreláció vizsgálata lakhely és tanulmányok között <a id="23"></a>
# [Tetejére! &#8673;](#home)

# In[102]:


df_live_stud=df.loc[:, ['livesin', 'studwhat']].copy()
df_live_stud.dropna(inplace=True)
df_live_stud = df_live_stud.groupby('studwhat').filter(lambda x : len (x) > 15)
df_live_stud = df_live_stud.groupby('livesin').filter(lambda x : len (x) >= 20)


# In[103]:


df_live_stud.iloc[:,0].size


# In[104]:


heatmap3 = pd.crosstab(df_live_stud.livesin, df_live_stud.studwhat)
heatmap3_norm_row=heatmap3.sub(heatmap3.mean(axis=1), axis=0)
heatmap3_norm_row = heatmap3_norm_row.div(heatmap3.std(axis=1), axis=0 )


# In[105]:


fig, ax = plt.subplots(figsize=(15,10))
ax = sns.heatmap(heatmap3, annot=True, cbar=False, fmt='g', annot_kws={"size": 14})
ax.set_title('Lakhely - tanulmányok', fontsize = 24, y =1.04)
ax.set_xlabel('Tanulmányok', fontsize=18);
ax.set_ylabel('Lakhely', fontsize=18);
plt.setp(ax.get_xticklabels(), rotation='vertical', fontsize=14);
plt.setp(ax.get_yticklabels(), rotation='horizontal', fontsize=14);


# Normalizálva sor zerint, mert Kolozsvár miatt a többi adat elenyészik.

# In[106]:


fig, ax = plt.subplots(figsize=(15,10))
ax = sns.heatmap(heatmap3_norm_row, annot=True, cbar=False, annot_kws={"size": 14})
ax.set_title('Lakhely - tanulmányok (sor szerint normalizálva)', fontsize=24, y=1.04)
ax.set_xlabel('Tanulmányok', fontsize=18);
ax.set_ylabel('Lakhely', fontsize=18);
plt.setp(ax.get_xticklabels(), rotation='vertical', fontsize=14);
plt.setp(ax.get_yticklabels(), rotation='horizontal', fontsize=14);


# ### 2.D. Korreláció vizsgálata lakhely és szakterület között <a id="24"></a>
# 
# [Tetejére! &#8673;](#home)

# In[107]:


df_live_work=df.loc[:, ['livesin', 'workwhat']].copy()
df_live_work.dropna(inplace=True)
df_live_work = df_live_work.groupby('livesin').filter(lambda x : len (x) >= 20)
df_live_work = df_live_work.groupby('workwhat').filter(lambda x : len (x) >= 10)


# In[108]:


heatmap4 = pd.crosstab(df_live_work.livesin, df_live_work.workwhat)
heatmap4_norm_row=heatmap4.sub(heatmap4.mean(axis=1), axis=0)
heatmap4_norm_row = heatmap4_norm_row.div(heatmap4.std(axis=1), axis=0 )


# In[109]:


fig, ax = plt.subplots(figsize=(15,10))
ax = sns.heatmap(heatmap4, annot=True, cbar=False, fmt='g', annot_kws={"size": 14})
ax.set_title('Lakhely - szakterület', fontsize = 24, y=1.04)
ax.set_xlabel('Szakterület', fontsize=18);
ax.set_ylabel('Lakhely', fontsize=18);
plt.setp(ax.get_xticklabels(), rotation='vertical', fontsize=14);
plt.setp(ax.get_yticklabels(), rotation='horizontal', fontsize=14);


# In[110]:


fig, ax = plt.subplots(figsize=(15,10))
ax = sns.heatmap(heatmap4_norm_row, annot=True, cbar=False, annot_kws={"size": 14})
ax.set_title('Lakhely - szakterület (sor szerint normalizálva)', fontsize = 24, y=1.04)
ax.set_xlabel('Szakterület', fontsize=18);
ax.set_ylabel('Lakhely', fontsize=18);
plt.setp(ax.get_xticklabels(), rotation='vertical', fontsize=14);
plt.setp(ax.get_yticklabels(), rotation='horizontal', fontsize=14);


# ### 2.E. Korreláció vizsgálata tanulmányok és szakterület között <a id="25"></a>
# 
# [Tetejére! &#8673;](#home)

# In[111]:


df_stud_work=df.loc[:, ['studwhat', 'workwhat']].copy()
df_stud_work.dropna(inplace=True)
df_stud_work = df_stud_work.groupby('studwhat').filter(lambda x : len (x) >= 15)
df_stud_work = df_stud_work.groupby('workwhat').filter(lambda x : len (x) >= 10)


# In[112]:


heatmap5 = pd.crosstab(df_stud_work.studwhat, df_stud_work.workwhat)
heatmap5_norm_row=heatmap5.sub(heatmap5.mean(axis=1), axis=0)
heatmap5_norm_row = heatmap5_norm_row.div(heatmap5.std(axis=1), axis=0 )


# In[113]:


fig, ax = plt.subplots(figsize=(15,10))
ax = sns.heatmap(heatmap5, annot=True, cbar=False, fmt='g', annot_kws={"size": 14})
ax.set_title('Tanulmányi terület - szakterület', fontsize = 24, y=1.04)
ax.set_xlabel('Szakterület', fontsize=18);
ax.set_ylabel('Tanulmányi terület', fontsize=18);
plt.setp(ax.get_xticklabels(), rotation='vertical', fontsize=14);
plt.setp(ax.get_yticklabels(), rotation='horizontal', fontsize=14);


# ## 3. Adatok mentése későbbi vizualizációhoz a [weboldalon](https://lorinczszabolcs.github.io/eletpalya) <a id="3"></a>
# 
# [Tetejére! &#8673;](#home)

# In[136]:


colors1 = np.array(['rgb(4,4,21)', 'rgb(55,15,79)',
          'rgb(105,27,127)', 'rgb(148,42,124)',
          'rgb(189,66,117)', 'rgb(230,108,102)',
          'rgb(251,144,103)', 'rgb(253,175,121)',
          'rgb(253,203,143)', 'rgb(252,230,168)',
          'rgb(255,255,255)']).tolist()

colors2 =np.array(['rgb(4,4,21)', 'rgb(35,11,56)',
                   'rgb(67,17,92)', 'rgb(98,24,127)',
                   'rgb(124,33,125)', 'rgb(150,43,124)',
                   'rgb(176,52,122)', 'rgb(201,78,113)',
                   'rgb(225,103,103)', 'rgb(250,129,94)',
                   'rgb(251,148,105)', 'rgb(253,166,116)',
                   'rgb(254,185,127)', 'rgb(253, 201, 142)',
                   'rgb(253, 218, 157)', 'rgb(252, 234, 172)', 
                   'rgb(252, 244, 220)', 'rgb(252, 255, 255)']).tolist()


# In[137]:


data_df=pd.crosstab(df_birth_stud.birth, df_birth_stud.studwhat)
data_df_normalized=pd.crosstab(df_birth_stud.birth, df_birth_stud.studwhat, normalize='index')
data={}
for i in range(len(data_df.columns)):
    x = np.array(data_df.index).tolist()
    y = np.array(data_df.iloc[:, i].values).tolist()
    name = str(data_df.columns[i])
    marker = dict([('color', colors1[i])])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/birth_stud.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)
data_normalized={}
for i in range(len(data_df_normalized.columns)):
    x = np.array(data_df_normalized.index).tolist()
    y = np.array(data_df_normalized.iloc[:, i].values).tolist()
    name = str(data_df_normalized.columns[i])
    marker = dict([('color', colors1[i])])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/birth_stud_normalized.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)

# heatmap

data = {}
x = np.array(data_df.columns).tolist()
y = np.array(data_df.index).tolist()
z = [[] for i in range(data_df.index.size)]

for i in range(len(data_df.index)):
    z[i] = np.array(data_df.iloc[i, :].values).tolist()
    
data[0] = dict([('x', x), ('y',y), ('z',z), ('type', 'heatmap'),
                ('colorscale', [[0, 'rgb(4, 4, 21)'],
                                [0.35, 'rgb(98,24,127)'],
                                [0.5, 'rgb(176,52,122)'],
                                [0.6, 'rgb(250, 129, 94)'],
                                [0.7, 'rgb(254,185,127)'],
                                [1, 'rgb(252, 234, 172)']
                               ]
                )
               ])

with open('../data/birth_stud_heatmap.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)


# In[138]:


data_df=pd.crosstab(df_birth_work.birth, df_birth_work.workwhat)
data_df_normalized=pd.crosstab(df_birth_work.birth, df_birth_work.workwhat, normalize='index')
data={}
for i in range(len(data_df.columns)):
    x = np.array(data_df.index).tolist()
    y = np.array(data_df.iloc[:, i].values).tolist()
    name = str(data_df.columns[i])
    marker = dict([('color', colors2[i])])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/birth_work.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)
data_normalized={}
for i in range(len(data_df_normalized.columns)):
    x = np.array(data_df_normalized.index).tolist()
    y = np.array(data_df_normalized.iloc[:, i].values).tolist()
    name = str(data_df_normalized.columns[i])
    marker = dict([('color', colors2[i])])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/birth_work_normalized.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)

# heatmap

data = {}
x = np.array(data_df.columns).tolist()
y = np.array(data_df.index).tolist()
z = [[] for i in range(data_df.index.size)]

for i in range(len(data_df.index)):
    z[i] = np.array(data_df.iloc[i, :].values).tolist()
    
data[0] = dict([('x', x), ('y',y), ('z',z), ('type', 'heatmap'),
                ('colorscale', [[0, 'rgb(4, 4, 21)'],
                                [0.35, 'rgb(98,24,127)'],
                                [0.5, 'rgb(176,52,122)'],
                                [0.6, 'rgb(250, 129, 94)'],
                                [0.7, 'rgb(254,185,127)'],
                                [1, 'rgb(252, 234, 172)']
                               ]
                )
               ])

with open('../data/birth_work_heatmap.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)


# In[139]:


data_df=pd.crosstab(df_live_stud.livesin, df_live_stud.studwhat)
data_df_normalized=pd.crosstab(df_live_stud.livesin, df_live_stud.studwhat, normalize='index')
data={}
for i in range(len(data_df.columns)):
    x = np.array(data_df.index).tolist()
    y = np.array(data_df.iloc[:, i].values).tolist()
    name = str(data_df.columns[i])
    marker = dict([('color', colors1[i])])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/live_stud.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)
data_normalized={}
for i in range(len(data_df_normalized.columns)):
    x = np.array(data_df_normalized.index).tolist()
    y = np.array(data_df_normalized.iloc[:, i].values).tolist()
    marker = dict([('color', colors1[i])])
    name = str(data_df_normalized.columns[i])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/live_stud_normalized.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)
    
# heatmap

data = {}
x = np.array(data_df.columns).tolist()
y = np.array(data_df.index).tolist()
z = [[] for i in range(data_df.index.size)]

for i in range(len(data_df.index)):
    z[i] = np.array(data_df.iloc[i, :].values).tolist()
    
data[0] = dict([('x', x), ('y',y), ('z',z), ('type', 'heatmap'),
                ('colorscale', [[0, 'rgb(4, 4, 21)'],
                                [0.35, 'rgb(98,24,127)'],
                                [0.5, 'rgb(176,52,122)'],
                                [0.6, 'rgb(250, 129, 94)'],
                                [0.7, 'rgb(254,185,127)'],
                                [1, 'rgb(252, 234, 172)']
                               ]
                )
               ])

with open('../data/live_stud_heatmap.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)


# In[140]:


data_df=pd.crosstab(df_live_work.livesin, df_live_work.workwhat)
data_df_normalized=pd.crosstab(df_live_work.livesin, df_live_work.workwhat, normalize='index')
data={}
for i in range(len(data_df.columns)):
    x = np.array(data_df.index).tolist()
    y = np.array(data_df.iloc[:, i].values).tolist()
    name = str(data_df.columns[i])
    marker = dict([('color', colors2[i])])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/live_work.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)
data_normalized={}
for i in range(len(data_df_normalized.columns)):
    x = np.array(data_df_normalized.index).tolist()
    y = np.array(data_df_normalized.iloc[:, i].values).tolist()
    marker = dict([('color', colors2[i])])
    name = str(data_df_normalized.columns[i])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/live_work_normalized.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)

# heatmap

data = {}
x = np.array(data_df.columns).tolist()
y = np.array(data_df.index).tolist()
z = [[] for i in range(data_df.index.size)]

for i in range(len(data_df.index)):
    z[i] = np.array(data_df.iloc[i, :].values).tolist()
    
data[0] = dict([('x', x), ('y',y), ('z',z), ('type', 'heatmap'),
                ('colorscale', [[0, 'rgb(4, 4, 21)'],
                                [0.35, 'rgb(98,24,127)'],
                                [0.5, 'rgb(176,52,122)'],
                                [0.6, 'rgb(250, 129, 94)'],
                                [0.7, 'rgb(254,185,127)'],
                                [1, 'rgb(252, 234, 172)']
                               ]
                )
               ])

with open('../data/live_work_heatmap.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)


# In[142]:


data_df=pd.crosstab(df_stud_work.studwhat, df_stud_work.workwhat)
data_df_normalized=pd.crosstab(df_stud_work.studwhat, df_stud_work.workwhat, normalize='index')
data={}
for i in range(len(data_df.columns)):
    x = np.array(data_df.index).tolist()
    y = np.array(data_df.iloc[:, i].values).tolist()
    name = str(data_df.columns[i])
    marker = dict([('color', colors2[i])])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/stud_work.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)
data_normalized={}
for i in range(len(data_df_normalized.columns)):
    x = np.array(data_df_normalized.index).tolist()
    y = np.array(data_df_normalized.iloc[:, i].values).tolist()
    name = str(data_df_normalized.columns[i])
    marker = dict([('color', colors2[i])])
    trace = dict([('x', x), ('y', y), ('name', name), ('marker', marker), ('type', 'bar')])
    data[i] = trace
with open('../data/stud_work_normalized.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)
    
# heatmap

data = {}
x = np.array(data_df.columns).tolist()
y = np.array(data_df.index).tolist()
z = [[] for i in range(data_df.index.size)]

for i in range(len(data_df.index)):
    z[i] = np.array(data_df.iloc[i, :].values).tolist()
    
data[0] = dict([('x', x), ('y',y), ('z',z), ('type', 'heatmap'),
                ('colorscale', [[0, 'rgb(4, 4, 21)'],
                                [0.35, 'rgb(98,24,127)'],
                                [0.5, 'rgb(176,52,122)'],
                                [0.6, 'rgb(250, 129, 94)'],
                                [0.7, 'rgb(254,185,127)'],
                                [1, 'rgb(252, 234, 172)']
                               ]
                )
               ])

with open('../data/stud_work_heatmap.json', 'w') as outfile:
    json.dump(data, outfile, ensure_ascii=True)


# In[ ]: