File size: 10,452 Bytes
42c6078 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 | #!/usr/bin/env python3
"""Generate all figures from processed results."""
import os
import sys
import json
import glob
import argparse
import pandas as pd
import numpy as np
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from src.plotting import (
plot_influence_vs_distance, plot_decay_ablation, plot_error_vs_radius,
plot_chi_vs_error, plot_interference_vs_chi, plot_runtime_vs_error,
plot_model_family_influence, plot_model_family_decay_mu,
plot_model_family_error_vs_radius, plot_model_family_proxy_vs_error,
plot_model_family_prior_noise_ablation
)
from src.utils import ensure_dir
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--config', type=str, default='config/default.yaml')
args = parser.parse_args()
proc_csv = 'results/processed/all_results.csv'
if not os.path.exists(proc_csv):
print(f"No processed results found at {proc_csv}. Run analyze_results.py first.")
return
df = pd.read_csv(proc_csv)
print(f"Loaded {len(df)} records")
fig_dir = ensure_dir('results/figures')
input_files = [proc_csv]
# Split
syn_df = df[df['dataset_type'] == 'synthetic'] if 'dataset_type' in df.columns else df
real_df = df[df['dataset_type'] == 'real'] if 'dataset_type' in df.columns else pd.DataFrame()
generated_figures = []
# ============================================================
# Synthetic figures
# ============================================================
if len(syn_df) > 0:
pg_df = syn_df[syn_df['model_family'] == 'poisson_gamma'] if 'model_family' in syn_df.columns else syn_df
if len(pg_df) > 0:
# S1: Influence vs distance
print("Generating Figure S1: influence vs distance...")
try:
paths = plot_influence_vs_distance(
pg_df, os.path.join(fig_dir, 'synthetic_influence_vs_distance.pdf'),
group_col='graph_type', title='Synthetic: Deletion Influence vs Distance',
input_files=input_files)
generated_figures.append(('S1', paths))
except Exception as e:
print(f" Error: {e}")
# S2: Decay ablation
print("Generating Figure S2: decay ablation...")
try:
paths = plot_decay_ablation(
pg_df, os.path.join(fig_dir, 'synthetic_decay_ablation.pdf'),
x_col='regime', y_col='empirical_decay_mu',
title='Synthetic: Empirical Decay Rate by Regime',
input_files=input_files)
generated_figures.append(('S2', paths))
except Exception as e:
print(f" Error: {e}")
# S3: Error vs radius
print("Generating Figure S3: error vs radius...")
try:
paths = plot_error_vs_radius(
pg_df, os.path.join(fig_dir, 'synthetic_error_vs_radius.pdf'),
group_col='graph_type', dataset_type='synthetic',
input_files=input_files)
generated_figures.append(('S3', paths))
except Exception as e:
print(f" Error: {e}")
# S4: chi vs error
print("Generating Figure S4: chi vs error...")
try:
paths = plot_chi_vs_error(
pg_df, os.path.join(fig_dir, 'synthetic_chi_vs_error.pdf'),
chi_col='chi_seed_max', error_col='rel_error_R2',
title='Synthetic: χ_max(z) vs Local Error (R=2)',
input_files=input_files)
generated_figures.append(('S4', paths))
except Exception as e:
print(f" Error: {e}")
# S5: Interference vs chi
print("Generating Figure S5: interference vs chi...")
try:
paths = plot_interference_vs_chi(
pg_df, os.path.join(fig_dir, 'synthetic_interference_vs_chi.pdf'),
chi_col='chi_seed_max', interf_col='interference_cosine_R2',
title='Synthetic: Interference vs χ_max(z)',
input_files=input_files)
generated_figures.append(('S5', paths))
except Exception as e:
print(f" Error: {e}")
# S6: Runtime vs error
print("Generating Figure S6: runtime vs error...")
try:
paths = plot_runtime_vs_error(
pg_df, os.path.join(fig_dir, 'synthetic_runtime_vs_error.pdf'),
title='Synthetic: Runtime vs Approximation Error',
input_files=input_files)
generated_figures.append(('S6', paths))
except Exception as e:
print(f" Error: {e}")
# ============================================================
# Real-data figures
# ============================================================
if len(real_df) > 0:
# R1: Influence vs distance
print("Generating Figure R1: real influence vs distance...")
try:
paths = plot_influence_vs_distance(
real_df, os.path.join(fig_dir, 'real_influence_vs_distance.pdf'),
group_col='dataset_name', title='Real Data: Deletion Influence vs Distance',
input_files=input_files)
generated_figures.append(('R1', paths))
except Exception as e:
print(f" Error: {e}")
# R2: Error vs radius
print("Generating Figure R2: real error vs radius...")
try:
paths = plot_error_vs_radius(
real_df, os.path.join(fig_dir, 'real_error_vs_radius.pdf'),
group_col='dataset_name', dataset_type='real',
input_files=input_files)
generated_figures.append(('R2', paths))
except Exception as e:
print(f" Error: {e}")
# R3: chi vs error
print("Generating Figure R3: real chi vs error...")
try:
paths = plot_chi_vs_error(
real_df, os.path.join(fig_dir, 'real_chi_vs_error.pdf'),
chi_col='chi_seed_max', error_col='rel_error_R2',
title='Real Data: χ_max(z) vs Local Error (R=2)',
input_files=input_files)
generated_figures.append(('R3', paths))
except Exception as e:
print(f" Error: {e}")
# R4: Runtime vs error
print("Generating Figure R4: real runtime vs error...")
try:
paths = plot_runtime_vs_error(
real_df, os.path.join(fig_dir, 'real_runtime_vs_error.pdf'),
title='Real Data: Runtime vs Approximation Error',
input_files=input_files)
generated_figures.append(('R4', paths))
except Exception as e:
print(f" Error: {e}")
# ============================================================
# Model-family figures
# ============================================================
if 'model_family' in df.columns and df['model_family'].nunique() > 1:
mf_df = df[df['dataset_type'] == 'synthetic'] if 'dataset_type' in df.columns else df
# M1: Influence by model family
print("Generating Figure M1: model family influence...")
try:
# Filter to bounded_degree for main figure
bd_df = mf_df[mf_df['graph_type'] == 'bounded_degree'] if 'graph_type' in mf_df.columns else mf_df
paths = plot_model_family_influence(
bd_df, os.path.join(fig_dir, 'model_family_influence_vs_distance.pdf'),
title='Influence Decay by Model Family (Bounded Degree)',
input_files=input_files)
generated_figures.append(('M1', paths))
except Exception as e:
print(f" Error: {e}")
# M2: Decay mu by model family
print("Generating Figure M2: model family decay mu...")
try:
# Need per-config aggregated mu_emp
mu_df = mf_df[mf_df['empirical_decay_mu'].notna()] if 'empirical_decay_mu' in mf_df.columns else pd.DataFrame()
if len(mu_df) > 0:
paths = plot_model_family_decay_mu(
mu_df, os.path.join(fig_dir, 'model_family_decay_mu.pdf'),
title='Empirical Decay Rate by Model Family',
input_files=input_files)
generated_figures.append(('M2', paths))
except Exception as e:
print(f" Error: {e}")
# M3: Error vs radius by model family
print("Generating Figure M3: model family error vs radius...")
try:
paths = plot_model_family_error_vs_radius(
mf_df, os.path.join(fig_dir, 'model_family_error_vs_radius.pdf'),
title='Error vs Radius by Model Family',
input_files=input_files)
generated_figures.append(('M3', paths))
except Exception as e:
print(f" Error: {e}")
# M4: Proxy vs error across families
print("Generating Figure M4: model family proxy vs error...")
try:
paths = plot_model_family_proxy_vs_error(
mf_df, os.path.join(fig_dir, 'model_family_proxy_vs_error.pdf'),
title='Interaction Proxy vs Error Across Models',
input_files=input_files)
generated_figures.append(('M4', paths))
except Exception as e:
print(f" Error: {e}")
# M5: Prior/noise ablation
print("Generating Figure M5: prior/noise ablation...")
try:
paths = plot_model_family_prior_noise_ablation(
mf_df, os.path.join(fig_dir, 'model_family_prior_noise_ablation.pdf'),
title='Prior/Noise Ablation by Model Family',
input_files=input_files)
generated_figures.append(('M5', paths))
except Exception as e:
print(f" Error: {e}")
print(f"\nGenerated {len(generated_figures)} figures:")
for name, paths in generated_figures:
print(f" {name}: {paths}")
if __name__ == '__main__':
main()
|