stacked_example_dan.py

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

input_directory = 'S:\E3 Projects\CEC Future of Nat Gas\PATHWAYS Model\Case Outputs\comb_outputs_20180629_1534 Draft Initial Scenarios'
output_directory = 'S:\E3 Projects\CEC Future of Nat Gas\PATHWAYS Model\Output Tools and Charts\python'
fmt = 'pdf'
cases = ['FONG High Electrification', 'FONG No Building Electrification']

outputs_path = output_directory

varname = 'Annual_Energy_Balanc'

gen_list = ['CHP', 'Coal', 'Imported Coal', 'Gas', 'Imported Gas', 'Imported BPA', 'Nuclear', 'Imported Nuclear', 'Large Hydro',
            'Other Renewables', 'Wind', 'Distributed PV', 'Utility PV', 'Curtailment']

yrange = [0, 800]
ylabel = 'Generation (TWh)'

gen_color_dict = {'Biomass': '#b3de69',
                 'CHP': 'grey',
                 'Coal': 'darkgrey',
                 'Distributed PV': '#fdb462',
                 'Imported Gas': '#d9d9d9',
                 'Large Hydro': '#80b1d3',
                 'Gas': 'silver',
                 'Nuclear': '#bebada',
                 'Other Renewables': '#fb8072',
                 'Utility PV': '#fed976',
                 'Wind': '#8dd3c7',
                 'Imported BPA' : 'dodgerblue',
                 'Imported Nuclear' : 'goldenrod',
                 'Curtailment' : 'r',
                 'Imported Coal' : 'slategrey'}

for case in cases:

    invar = pd.read_csv(os.path.join(input_directory, varname + '.csv'))
    var = invar[invar.Active_Cases == case]

    pivot = pd.crosstab(var['Output_Year'], var['Energy_Balance_Index'], values=var['Value'], aggfunc=np.sum)
    pivot[np.isnan(pivot)] = 0.
    gen = pivot[gen_list]

    gen = gen.applymap(lambda x: 0 if x < 0 else x)
    
    fig, ax = plt.subplots() ## this is the most flexible approach to access the mpl api
                             ## can easily do subplots this way 
    #plt.figure()
    
    gen.plot.area(ax = ax, fontsize = 11, 
                  color = [gen_color_dict[x] for x in gen.columns.values]) ## list comprehension using color dict above
    #plt.stackplot(gen.index, gen.values.transpose())
    
    ax.set_ylim(yrange) 
    #plt.ylim(yrange)
    
    ax.set_ylabel(ylabel)
    #plt.ylabel(ylabel)

    ax.set_xlim([2015, 2050])
    
    ax.set_xlabel('')
    
    
    
    handles, labels = ax.get_legend_handles_labels() ## get legend labels and boxes as variables
    lgd = ax.legend(handles = handles[::-1], labels = labels[::-1], bbox_to_anchor = [1,1]) ## reverse order
                                                                                            ## bbox moves the legend in more precise fashion
    #plt.legend(gen.columns, loc='upper left', bbox_to_anchor = [1, 1])
    
    ax.set_title(case, fontsize = 14)
    #plt.title(case)
    
    ## I usually save as a png 
    ## the legend will be cut off unless use do the bbox extra artists 

    plt.savefig(outputs_path + '//' + varname + '_' + case + '.png', 
            dpi = 600, transparent = True, bbox_extra_artists = (lgd,),
            bbox_inches = 'tight')
    # plt.savefig(os.path.join(output_directory, varname + ', '  + case + '.' + fmt), format=fmt,
    #             )
    
    plt.show()