GenderNetworks_Repository/variable_selection.py at main · MemoOlv/GenderNetworks_Repository · 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
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
#! /usr/bin/python

import sys
import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
import seaborn as sns
import typer

import lib.data_processing_functions as dpf

app = typer.Typer()


@app.command()
def variable_selection(year: int):
    lib_path = "/lib"
    sys.path.insert(0, lib_path)

    sns.set_style("whitegrid", {"grid.linestyle": "--", "axes.edgecolor": "0"})
    data_path = "data/"
    yr = str(year)

    # Read ENIGH clean data
    ENIGHr = pd.read_csv(data_path + "ENIGH" + yr + "_clean.csv", index_col=[0])
    ENIGH = ENIGHr.copy()

    # Add columns to ENIGH for data classification
    t = "CH"  # Options CH,CP,MEPI
    ENIGH = dpf.hhld_classification(ENIGH)

    # Computation of the representativity
    # 0->cov_matrix 1->keys, 2>represent, 3->n_nodes
    R = dpf.ComputeRepresentativity(ENIGH, t, yr, data_path)

    # frames0 is a list that contains the criteria selection
    frames0 = list()
    for i in range(0, len(R[2])):
        frames0.append(dpf.criteria(R[2], R[1][i], R[3])[0])
    # frames1 is a dictionary that contains the covariance data
    frames1 = {}
    for i in range(0, len(R[2])):
        frames1[R[1][i]] = dpf.criteria(R[2], R[1][i], R[3])[1]
    crt = pd.concat(frames0, ignore_index=True)

    k = R[1][0]  # Select the person classification to plot
    crt_p = frames1[k]
    crt_pmax = crt_p[crt_p["covmul"] == crt_p["covmul"].max()]
    m = crt.value.mean()
    crt_sort = crt.sort_values(by="num")

    f = plt.figure(figsize=(12, 8))
    f.subplots_adjust(hspace=0.0)
    fontsize_ticks = 20
    gs = gridspec.GridSpec(1, 1)
    ax = plt.subplot(gs[0])
    ax.tick_params(labelsize=fontsize_ticks)
    sns.despine()
    sns.scatterplot(data=crt_p, x="cov", y="covmul", alpha=0.8, size="covmul", hue="covmul", ax=ax)
    sns.scatterplot(data=crt_pmax, x="cov", y="covmul", color="red", ax=ax, label="$x_{m}$")
    ax.legend(loc="best", frameon=True, fontsize=(fontsize_ticks - 5), title="Rep*Cov")
    plt.xlabel("Cov", fontsize=fontsize_ticks)
    plt.ylabel("Rep*Cov", fontsize=fontsize_ticks)
    plt.savefig("reports/figures/RepCov_HAdultos" + yr + ".pdf", bbox_inches="tight")
    plt.show()

    f = plt.figure(figsize=(8 * 1.6, 8))
    f.subplots_adjust(hspace=0.0)
    fontsize_ticks = 20
    gs = gridspec.GridSpec(1, 1)
    ax = plt.subplot(gs[0])
    ax.tick_params(labelsize=fontsize_ticks)
    sns.despine()
    sns.scatterplot(data=crt_p, x="idx", y=k, alpha=0.8, size="cov", hue="cov", ax=ax)
    # sns.scatterplot(data=crt_pmax, x='idx', y=k, color='red',ax=ax)
    for line in range(0, 13):
        n_s = dpf.recur_fibo(line)
        plt.text(crt_p.idx[n_s], crt_p[k][n_s], crt_p.index[n_s], horizontalalignment="center")
    plt.xlabel("No. Variable", fontsize=fontsize_ticks)
    plt.ylabel("Rep", fontsize=fontsize_ticks)
    ax.legend(loc="best", frameon=True, fontsize=(fontsize_ticks - 5), title="Cov")
    plt.yscale("log")
    plt.xscale("log")
    plt.savefig("reports/figures/Representatividad_HAdultos" + yr + ".pdf", bbox_inches="tight")
    plt.show()

    f = plt.figure(figsize=(10 * 1.618, 8))
    f.subplots_adjust(hspace=0.0)
    fontsize_ticks = 20
    gs = gridspec.GridSpec(1, 1)
    ax = plt.subplot(gs[0])
    ax.tick_params("y", labelsize=fontsize_ticks)
    sns.despine()
    new_ticks = [i for i in crt_sort.PC]
    plt.xticks(
        range(0, len(new_ticks), 3), new_ticks[::3], fontsize=(fontsize_ticks - 5), rotation=90
    )
    ax = sns.lineplot(data=crt_sort, x="PC", y="value", marker="o", label="Rep$_{m}$")
    ax2 = ax.twinx()
    ax2.tick_params(labelsize=fontsize_ticks)
    sns.lineplot(
        data=crt_sort, x="PC", y="num", ax=ax2, color="green", marker="o", label="No. personas"
    )
    ax.axhline(y=m, linestyle="--", color="darkblue")
    ax.set_ylabel("Rep", fontsize=fontsize_ticks)
    ax.set_xlabel("", fontsize=fontsize_ticks)
    ax2.set_ylabel("Número de personas", fontsize=fontsize_ticks)
    ax.set_ylim((0, 1.05))
    ax2.set_ylim((0, 100000))
    ax.legend(loc="upper left", frameon=True)
    plt.savefig("reports/figures/Rep_Pob_Hogares" + yr + ".pdf", bbox_inches="tight")
    plt.show()

    # print(R[1])

    # print(frames1)
    cov_matrix_cut = dpf.typeofcriteria(R[1], "static", frames1, m, crt)
    # cov_matrix_cut.drop('foliohog',inplace=True)

    f = plt.figure(figsize=(30, 30))
    f.subplots_adjust(hspace=0.0)
    fontsize_ticks = 25
    gs = gridspec.GridSpec(1, 1)
    ax = plt.subplot(gs[0])
    ax.tick_params(labelsize=fontsize_ticks)
    sns.heatmap(cov_matrix_cut, cmap="jet")
    cax = plt.gcf().axes[-1]
    cax.tick_params(labelsize=fontsize_ticks)
    plt.savefig("reports/figures/CovMatrixCut" + yr + ".pdf", bbox_inches="tight")
    plt.show()

    cov_matrix_cut.to_csv(data_path + "cov_matrix_" + t + yr + "_cut.csv", index=True)


if __name__ == "__main__":
    typer.run(variable_selection)