SR-ARPOD/Plots
2026-04-01 23:32:10 +08:00
..
data chore: save current workspace state 2026-04-01 23:32:10 +08:00
__init__.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.1_eval_success_qp_rates.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.1_eval_success_qp_rates.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.2_eval_mean_return.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.2_eval_mean_return.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.3_train_entropy.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.3_train_entropy.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.4_train_intervention_rate.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.4_train_intervention_rate.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.5_train_reward.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
1.5_train_reward.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig1_training_curves.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig1_training_curves.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig1_training_curves.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig2_trajectory_3d_interactive_full50.html chore: save current workspace state 2026-04-01 23:32:10 +08:00
fig2_trajectory_3d_interactive.html chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig2_trajectory_3d_interactive.py chore: save current workspace state 2026-04-01 23:32:10 +08:00
fig2_trajectory_3d_v2.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig2_trajectory_3d.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig2_trajectory_3d.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig2_trajectory_3d.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig3_state_convergence.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig3_state_convergence.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig3_state_convergence.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig4_control_decomposition.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig4_control_decomposition.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig4_control_decomposition.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig5_safety_analysis.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig5_safety_analysis.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig5_safety_analysis.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig6_ablation.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig6_ablation.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig6_ablation.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig7_monte_carlo.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig7_monte_carlo.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig7_monte_carlo.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig8_error_convergence.pdf chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig8_error_convergence.png chore: save current workspace state 2026-04-01 22:48:53 +08:00
fig8_error_convergence.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
generate_all.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
generate_plot_from_checkpoint.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
parse_training_logs.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
plot_config.py chore: save current workspace state 2026-04-01 22:48:53 +08:00
README.md chore: save current workspace state 2026-04-01 22:48:53 +08:00
run_evaluation.py chore: save current workspace state 2026-04-01 23:32:10 +08:00

STELLAR 论文图表生成

本目录包含用于生成期刊论文图表的 Python 脚本。

快速开始

# 1. 解析训练日志 → CSV
python -m Plots.parse_training_logs

# 2. 运行模型评估 → .npz 轨迹数据
python -m Plots.run_evaluation --n_episodes 50

# 3. 生成全部图表
python -m Plots.generate_all --plot_only

# 或一键完成全部流程
python -m Plots.generate_all --full

图表清单

编号 文件 内容 数据来源
图 1 fig1_training_curves.py 训练收敛(成功率、回报、熵、介入率) CSV / TB CSV
图 2 fig2_trajectory_3d.py 3D 轨迹 + 安全约束可视化 .npz
图 3 fig3_state_convergence.py 位置/速度时间历程 .npz
图 4 fig4_control_decomposition.py 控制分解u_nom, u_res, u_applied .npz
图 5 fig5_safety_analysis.py HOCBF 约束值 (h_c, h_a, h_) .npz
图 6 fig6_ablation.py 消融实验对比柱状图 消融评估
图 7 fig7_monte_carlo.py 蒙特卡洛统计(误差散点/分布/奖励) .npz
图 8 fig8_error_convergence.py 误差收敛包络 .npz

基础设施

  • plot_config.py — 全局样式(字体、颜色、尺寸)
  • parse_training_logs.py — 训练日志解析(多阶段文本 + TensorBoard
  • run_evaluation.py — 模型评估与轨迹数据采集
  • generate_all.py — 一键流水线

数据位置

  • 训练日志: Logs/contv3_hybrid40h_v2_20260316_140251/train.log
  • 最佳检查点: Checkpoint/contv3_hybrid40h_v2_20260316_140251/phase2/best_model.pt
  • TB 导出 CSV: Plots/tb_exports/tb_scalars_all.csv
  • 中间数据: Plots/data/ (解析后的 CSV 和评估 .npz)

输出格式

每张图同时输出 PDF矢量和 PNG300 dpi保存于 Plots/ 根目录。