grapher.py

import pandas as pd
import numpy as np
import migrant_functions as mf
import qgrid as qg
import matplotlib.pyplot as plt
import matplotlib as mtplt
import jordan as j
import random
from matplotlib import gridspec
import sys
sys.path.insert(0, "development")

# Takes one country, and compares different aspects of it by gender (and possibly by year)
# Arguemnts:
    # country: Country of Interest
    # aspect: Numerical [1, 2, 3]
# Optional Argument(s):
    # Year (Defaults to False): If True, displays a cluster of bar graphs for each year
# Possible Aspects Key:
    # Aspect 1: Immgration
    # Aspect 2: Emigration
    # Aspect 3: Age Ranges (Immigration Only)

migrant_data = pd.read_csv('development/migrant_table_final.csv').drop(columns=['Unnamed: 0'])

def one_pick_and_graph(country, aspect, year=False):

    conversion = {1:'Immigration', 2:'Emigration', 3:'Age Ranges (Immigration)'}


    if aspect not in [1, 2, 3]:
        print('Not a valid aspect to compare!')
        return

    if aspect == 1:
        selected_data = migrant_data.loc[(migrant_data['Country'] == country)
                                     & (migrant_data['Migration Type'] == 'Immigration'),:]

        if year:
            picks = selected_data[['Year', 'Gender', 'Total Migration']]
        else:
            picks = selected_data[['Gender', 'Total Migration']].groupby('Gender').agg(sum)

    elif aspect == 2:
        selected_data = migrant_data.loc[(migrant_data['Country'] == country)
                                     & (migrant_data['Migration Type'] == 'Emigration'),:]

        if year:
            picks = selected_data[['Year', 'Gender', 'Total Migration']]
        else:
            picks = selected_data[['Gender', 'Total Migration']].groupby('Gender').agg(sum)

    elif aspect == 3:
        selected_data = migrant_data.loc[(migrant_data['Country'] == country)
                                     & (migrant_data['Migration Type'] == 'Immigration'),:]
        if year:
            picks = selected_data[['Year', 'Gender', 'Migrants Under 15 years old',
                                   'Migrants 20-29 years old',
                                   'Migrants 30-39 years old',
                                   'Migrants 40-49 years old',
                                   'Migrants 50 years old and older']]
        else:
            picks = selected_data[['Gender', 'Migrants Under 15 years old',
                                   'Migrants 20-29 years old',
                                   'Migrants 30-39 years old',
                                   'Migrants 40-49 years old',
                                   'Migrants 50 years old and older']].groupby('Gender').agg(sum)


    if not year:
        plotty = picks.plot(kind='bar', figsize=(7,7));
        plt.xlabel('Gender', size=15)
        plt.ylabel('Counts', size=15)
        plt.title(conversion[aspect] + ' Counts', size=15)
    else:
        if aspect != 3:

            barWidth = 0.25

            # set height of bar
            bars = []
            for gender in ['female', 'male', 'total']:
                picks1 = picks.loc[picks['Gender'] == gender,:]
                bars1 = list(picks1.loc[:,'Total Migration'].values)
                bars.append(bars1)


            # Set position of bar on X axis
            current = np.arange(len(bars[0]))
            positions = [current]
            for bar in bars[1:]:
                r = [x + barWidth for x in current]
                positions.append(r)
                current = r

            #Make the plot
            for pos, bar, label in zip(positions, bars, ['female', 'male', 'total']):
                plt.bar(pos, bar, color=(random.uniform(0, 1), random.uniform(0, 1), random.uniform(0, 1)), width=barWidth, edgecolor='white', label=label)

            # Add xticks on the middle of the group bars
            plt.xlabel('Year', size=15)
            plt.xlabel('Counts', size=15)
            plt.title(conversion[aspect] + ' Counts by Year', size=15)
            plt.xticks([r + barWidth for r in range(len(bars[0]))], ['1990', '1995', '2000',
                                                                     '2005', '2010', '2015', '2017'])
            plt.xlabel('Gender', size=15)
            plt.ylabel('Counts', size=15)
            fig_size = plt.rcParams["figure.figsize"]
            fig_size[0] = 7
            fig_size[1] = 7
            plt.legend()

        # CREATE SUBPLOTS
        else:

            fig1, axes = plt.subplots(nrows=1, ncols=2, figsize=(30,15));

            axes[0].set_title("Age Range Immigration by Year, Male", size=20)
            axes[0].set_ylabel('Counts', size=22)
            axes[0].set_xlabel('Year', size=22)
            axes[1].set_xlabel('Year', size=22)
            axes[1].set_title("Age Range Immigration by Year, Female", size=20)


            plt.legend(prop={'size': 6})


            for tick in axes[0].xaxis.get_major_ticks():
                tick.label.set_fontsize(22)

            for tick in axes[1].xaxis.get_major_ticks():
                tick.label.set_fontsize(22)

            for tick in axes[0].yaxis.get_major_ticks():
                tick.label.set_fontsize(22)


            male = picks.loc[picks['Gender'] == 'male'].drop(columns='Gender')
            m = male.plot(x='Year', kind='line', ax=axes[0]);
            m.legend(prop={'size': 15})

            female = picks.loc[picks['Gender'] == 'female'].drop(columns='Gender')
            f = female.plot(x='Year', kind='line', ax=axes[1], sharey = m);
            f.legend(prop={'size': 15})

            total = picks.loc[picks['Gender'] == 'total'].drop(columns='Gender')
            t = total.plot(x='Year', kind='line', figsize=(25,15));
            t.legend(prop={'size': 15})
            t.set_ylabel('Counts', size=22)
            t.set_xlabel('Year', size=22)
            t.set_title("Age Range Immigration by Year, Total", size=22)

            for tick in t.xaxis.get_major_ticks():
                tick.label.set_fontsize(22)

            for tick in t.yaxis.get_major_ticks():
                tick.label.set_fontsize(22)

# Compares Multiple Countries
# Arguments:
    # variable: Gender
    # variable_type: Male or Female
    # immigraiton_type: type of migration (Immigration or Emigration)
        # this is to be specified when looking at gender as the variable
    # interest_country_list: countries to compare
    # plot_type: 'line' or 'bar'

# either a bar graph or line graph
    # if bar, x = year, y=variable
    # if line, y=country, x=year

def multiple_pick_and_graph(migrant_data, variable, variable_type, immigration_type, interest_country_list, plot_type):
    selected_data = migrant_data.loc[(migrant_data[variable] == variable_type)
                                     & (migrant_data['Migration Type'] == immigration_type),:]

    picks = selected_data[['Country', 'Year', 'Total Migration']]

    if plot_type == 'line':

        plt.figure(figsize=(10,10))

        plt.style.use('seaborn-white')
#         my_dpi=96
#         plt.figure(figsize=(480/my_dpi, 480/my_dpi), dpi=my_dpi)

        region =['Cuba',
         'El Salvador',
         'Guatemala',
         'Honduras',
         'Mexico',
         'Venezuela (Bolivarian Republic of)']

        # multiple line plot
        country_list = list(np.unique(picks['Country']))
        end_values = []
        for country in country_list:
            country_picks = picks.loc[picks['Country'] == country,:]
            if country != interest_country_list[0]:
                plt.plot(country_picks['Year'].values, np.log(country_picks['Total Migration'].values), marker='', color='black', linewidth=1, alpha=0.4)
            else:
                end_value = picks.loc[picks['Country'] == country,:].values[6][2]
                plt.plot(country_picks['Year'].values, np.log(country_picks['Total Migration'].values), marker='', color='blue', linewidth=5, alpha=0.7)

        plt.xlim(1989, 2018);
        plt.ylim(7, 20);

        def rand_jitter(arr):
            stdev = .01*(max(arr)-min(arr))
            return arr + np.random.randn(len(arr)) * stdev



        plt.text(2019, np.log(end_value), interest_country_list[0], horizontalalignment='left', size='large', color='black')

        plt.title('Female' + ' ' + 'immigration ' + "For Each Country, by Year", loc='left', fontsize=15, fontweight=0, color='black');
        plt.xlabel("Year", size=15);
        plt.ylabel("log(Immigration)", size=15);
        plt.show()


    elif plot_type == 'bar':

        plt.figure(figsize=(10,10))

        # set width of bar
        barWidth = 0.25

        # set height of bar
        bars = []
        for country in interest_country_list:
            picks1 = picks.loc[picks['Country'] == country,:]
            bars1 = list(picks1.loc[:,'Total Migration'].values)
            bars.append(bars1)


        # Set position of bar on X axis
        current = np.arange(len(bars[0]))
        positions = [current]
        for bar in bars[1:]:
            r = [x + barWidth for x in current]
            positions.append(r)
            current = r

        #Make the plot
        for pos, bar, label in zip(positions, bars, interest_country_list):
            plt.bar(pos, bar, color=(random.uniform(0, 1), random.uniform(0, 1), random.uniform(0, 1)), width=barWidth, edgecolor='white', label=label)

        # Add xticks on the middle of the group bars
        plt.title('Side by Side Comprison of ' + immigration_type + ' for Selected Countries, by Year', size=15)
        plt.xlabel('Year', fontweight='bold', size=15)
        plt.ylabel('Counts', fontweight='bold', size=15)
        plt.xticks([r + barWidth for r in range(len(bars[0]))], ['1990', '1995', '2000',
                                                                 '2005', '2010', '2015', '2017'])


        # Create legend & Show graphic
        plt.legend(prop={'size': 15})
        plt.show()

    else:
        print('Not a valid visulization!')