Parametros_de_secciones/tratar datos/con todos datos.py

import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
import numpy as np
import warnings
warnings.filterwarnings('ignore')

# ── Leer datos ──────────────────────────────────────────────────────────────
df = pd.read_csv(
    'resultados1.txt',
    sep=r'\s+',
    decimal=',',
    header=0
)

# Variables de interés
inputs  = ['H', 'b', 'tf', 'tw', 'e', 'L']
outputs = ['Flecha_Media', 'Peso']

# Convertir a numérico por si acaso
for col in inputs + outputs:
    df[col] = pd.to_numeric(df[col], errors='coerce')

df.dropna(subset=inputs + outputs, inplace=True)

print(f"Filas cargadas: {len(df)}")
print(df[inputs + outputs].describe().to_string())

# ── Paleta de colores ────────────────────────────────────────────────────────
COLORS = {
    'H':  '#E63946',
    'b':  '#457B9D',
    'tf': '#2A9D8F',
    'tw': '#E9C46A',
    'e':  '#F4A261',
    'L':  '#A8DADC',
}
BG   = '#0F1117'
GRID = '#2A2D3A'
TEXT = '#E8EAF6'

plt.rcParams.update({
    'figure.facecolor': BG,
    'axes.facecolor':   BG,
    'axes.edgecolor':   GRID,
    'axes.labelcolor':  TEXT,
    'xtick.color':      TEXT,
    'ytick.color':      TEXT,
    'grid.color':       GRID,
    'text.color':       TEXT,
    'font.family':      'monospace',
})

labels = {
    'H':  'H (altura)',
    'b':  'b (ancho)',
    'tf': 'tf (ala)',
    'tw': 'tw (alma)',
    'e':  'e (excentr.)',
    'L':  'L (longitud)',
}

# ════════════════════════════════════════════════════════════════════════════
# FIGURA 1 – Scatter: cada entrada vs Flecha_Media y Peso
# ════════════════════════════════════════════════════════════════════════════
fig1, axes = plt.subplots(
    2, 6, figsize=(22, 8),
    facecolor=BG,
    gridspec_kw={'hspace': 0.45, 'wspace': 0.35}
)
fig1.suptitle('Entradas vs Salidas  ·  Scatter', fontsize=15, fontweight='bold',
              color=TEXT, y=1.01)

for col_idx, inp in enumerate(inputs):
    for row_idx, out in enumerate(outputs):
        ax = axes[row_idx, col_idx]
        color = COLORS[inp]

        # Muestreo para no saturar
        sample = df[[inp, out]].dropna().sample(min(2000, len(df)), random_state=42)

        ax.scatter(sample[inp], sample[out],
                   s=6, alpha=0.35, color=color, linewidths=0)

        # Línea de tendencia
        z = np.polyfit(sample[inp], sample[out], 1)
        p = np.poly1d(z)
        xs = np.linspace(sample[inp].min(), sample[inp].max(), 200)
        ax.plot(xs, p(xs), color='white', lw=1.2, alpha=0.7)

        if out == 'Flecha_Media':
            ax.axhline(-25, color='white', lw=1.2, linestyle='--', alpha=0.85, label='-25')
            ax.legend(fontsize=7, framealpha=0.2, labelcolor=TEXT)

        ax.set_xlabel(labels[inp], fontsize=8)
        ax.set_ylabel(out, fontsize=8)
        ax.grid(True, linestyle='--', alpha=0.3)
        ax.tick_params(labelsize=7)

fig1.tight_layout()
fig1.savefig('scatter_entradas_vs_salidas.png',
             dpi=150, bbox_inches='tight', facecolor=BG)
print("✓ scatter_entradas_vs_salidas.png guardado")

# ════════════════════════════════════════════════════════════════════════════
# FIGURA 2 – Distribuciones (histogramas) de entradas + salidas
# ════════════════════════════════════════════════════════════════════════════
all_vars = inputs + outputs
n = len(all_vars)
fig2, axes2 = plt.subplots(2, 4, figsize=(18, 8), facecolor=BG,
                            gridspec_kw={'hspace': 0.5, 'wspace': 0.35})
fig2.suptitle('Distribución de variables', fontsize=15, fontweight='bold',
              color=TEXT, y=1.01)
axes2_flat = axes2.flatten()

for i, var in enumerate(all_vars):
    ax = axes2_flat[i]
    clr = COLORS.get(var, '#90CAF9')
    data = df[var].dropna()
    ax.hist(data, bins=40, color=clr, alpha=0.85, edgecolor='none')
    ax.axvline(data.mean(), color='white', lw=1.4, linestyle='--', label=f'μ={data.mean():.3g}')
    ax.set_title(var, fontsize=10, fontweight='bold', color=clr)
    ax.set_ylabel('Frecuencia', fontsize=8)
    ax.grid(True, linestyle='--', alpha=0.3)
    ax.tick_params(labelsize=7)
    ax.legend(fontsize=7, framealpha=0.2, labelcolor=TEXT)

# Ocultar último subplot vacío si n < 8
for j in range(n, len(axes2_flat)):
    axes2_flat[j].set_visible(False)

fig2.tight_layout()
fig2.savefig('distribuciones.png',
             dpi=150, bbox_inches='tight', facecolor=BG)
print("✓ distribuciones.png guardado")

# ════════════════════════════════════════════════════════════════════════════
# FIGURA 3 – Correlación: heatmap
# ════════════════════════════════════════════════════════════════════════════
corr = df[inputs + outputs].corr()

fig3, ax3 = plt.subplots(figsize=(9, 7), facecolor=BG)
fig3.suptitle('Mapa de correlación', fontsize=14, fontweight='bold', color=TEXT)

cmap = plt.cm.RdYlGn
im = ax3.imshow(corr.values, cmap=cmap, vmin=-1, vmax=1, aspect='auto')
plt.colorbar(im, ax=ax3, fraction=0.046, pad=0.04).ax.tick_params(colors=TEXT)

ticks = list(range(len(corr.columns)))
ax3.set_xticks(ticks); ax3.set_yticks(ticks)
ax3.set_xticklabels(corr.columns, rotation=45, ha='right', fontsize=9)
ax3.set_yticklabels(corr.columns, fontsize=9)

for i in range(len(corr)):
    for j in range(len(corr)):
        val = corr.values[i, j]
        color_txt = 'black' if abs(val) > 0.5 else TEXT
        ax3.text(j, i, f'{val:.2f}', ha='center', va='center',
                 fontsize=8, color=color_txt, fontweight='bold')

ax3.grid(False)
fig3.tight_layout()
fig3.savefig('correlacion_heatmap.png',
             dpi=150, bbox_inches='tight', facecolor=BG)
print("✓ correlacion_heatmap.png guardado")

# ════════════════════════════════════════════════════════════════════════════
# FIGURA 4 – Boxplots de salidas por cuantiles de L
# ════════════════════════════════════════════════════════════════════════════
df['L_cat'] = pd.qcut(df['L'], q=5, labels=['L_Q1','L_Q2','L_Q3','L_Q4','L_Q5'])

fig4, axes4 = plt.subplots(1, 2, figsize=(14, 6), facecolor=BG)
fig4.suptitle('Distribución de salidas por cuantiles de L', fontsize=14,
              fontweight='bold', color=TEXT)

palette = ['#E63946','#F4A261','#E9C46A','#2A9D8F','#457B9D']

for ax_idx, out in enumerate(outputs):
    ax = axes4[ax_idx]
    groups = [df.loc[df['L_cat'] == cat, out].dropna().values
              for cat in df['L_cat'].cat.categories]

    bp = ax.boxplot(groups, patch_artist=True, notch=False,
                    medianprops=dict(color='white', lw=2),
                    whiskerprops=dict(color=TEXT),
                    capprops=dict(color=TEXT),
                    flierprops=dict(marker='o', color=TEXT, alpha=0.2, markersize=2))

    for patch, clr in zip(bp['boxes'], palette):
        patch.set_facecolor(clr)
        patch.set_alpha(0.75)

    ax.set_xticklabels(df['L_cat'].cat.categories, rotation=20, fontsize=8)
    ax.set_title(out, fontsize=11, fontweight='bold', color=TEXT)
    ax.set_ylabel(out, fontsize=9)
    ax.grid(True, axis='y', linestyle='--', alpha=0.3)

fig4.tight_layout()
fig4.savefig('boxplot_salidas_por_L.png',
             dpi=150, bbox_inches='tight', facecolor=BG)
print("✓ boxplot_salidas_por_L.png guardado")

print("\n✅ Todos los gráficos generados correctamente.")