PythonClassIpea/graficos_real2.py at master · BAFurtado/PythonClassIpea · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
import matplotlib.pyplot as plt
import pandas as pd
from pylab import Circle

import means_comparison
from data import groups_cols

colors = ['tab:blue', 'tab:red']


def plotting(data, different_means, col1='z_simulated_optimal', col2='z_ml_optimal', name='name'):
    lbsl = [groups_cols.abm_params_show[l] for l in data.index.tolist() if l in different_means]
    fig, ax = plt.subplots(figsize=(8, 3))
    for par in data.index:
        if par in different_means:
            plt.hlines(par,
                       xmin=min(min(d[col1]), min(d[col2])),
                       xmax=max(max(d[col1]), max(d[col2])),
                       colors='lightgrey', lw=.8, alpha=.5)
            ax.plot([data.loc[par, col1], data.loc[par, col2]], [par, par],
                    color=colors[0] if data.loc[par, col1] > data.loc[par, col2]
                    else colors[1], lw=1, alpha=.8)
            for i, each in enumerate([col1, col2]):
                ax.scatter(data.loc[par, each], par, color=colors[i], alpha=.9, marker='o')

    plt.yticks(range(len(lbsl)), lbsl, fontsize=10)

    pts = Circle((0, 0))
    ax.legend([pts, pts], ['ABM Simulated optimal', 'ML surrogate optimal'],
              edgecolor='white', facecolor='white',
              framealpha=1, loc=(-.6, -.3))
    # plt.ylabel("Model parameters")
    plt.xlabel(f"Mean value of standardized parameters for optimal results")
    leg = ax.get_legend()
    for i in range(len(colors)):
        # leg.legendHandles[i].set_marker('o')
        leg.legendHandles[i].set_color(colors[i])
    plt.tight_layout()
    plt.savefig(f'parameters.png', dpi=300)
    plt.show()


if __name__ == '__main__':
    o = pd.read_csv('data/means_comparison_output_normTrue.csv', sep=';')
    o.rename(columns={'Unnamed: 0': 'param'}, inplace=True)
    o2 = pd.read_csv('data/means_comparison_optimal_non_optimal_normTrue.csv', sep=';')
    o2.rename(columns={'Unnamed: 0': 'param'}, inplace=True)
    lst_parameters = means_comparison.different(o, o2)

    d = pd.read_csv('data/parameters_norm_optimal.csv', sep=';')
    d.rename(columns={'Unnamed: 0': 'parameters'}, inplace=True)
    d = d.set_index('parameters')
    d = d.sort_values(by='difference')
    plotting(d, lst_parameters)