feat: 添加强化学习项目报告及重构课程作业报告代码结构

- 新增强化学习个人项目报告，包含基于PyTorch从零实现的PPO算法 - 重构课程作业报告代码结构，提取运行时路径管理和notebook执行逻辑到独立模块 - 更新依赖文件requirements.txt，添加强化学习相关依赖 - 简化模型比较结果表格，仅保留基线逻辑回归模型数据
2026-04-30 16:54:41 +08:00
parent 6ac02ba4fe
commit d353133b31
21 changed files with 1639 additions and 102 deletions
@@ -0,0 +1,55 @@
+import json
+import traceback
+from pathlib import Path
+
+from .runtime_paths import build_paths
+
+
+def execute_notebook(
+    start_at: int | None = None,
+    stop_at: int | None = None,
+    namespace: dict | None = None,
+) -> dict:
+    paths = build_paths()
+    paths.ensure_outputs()
+
+    nb_data = json.loads(paths.notebook.read_text(encoding="utf-8"))
+    code_cells = [c for c in nb_data["cells"] if c["cell_type"] == "code"]
+    if not code_cells:
+        return {"status": "skipped", "reason": "no code cells found"}
+
+    ns = (namespace or {}).copy()
+    ns.update(paths.as_injection())
+    ns["RANDOM_STATE"] = 42
+
+    start = max((start_at or 1) - 1, 0)
+    stop = stop_at if stop_at is not None else len(code_cells)
+    cells_to_run = code_cells[start:stop]
+
+    results = []
+    for i, cell in enumerate(cells_to_run, start=start + 1):
+        src = "".join(cell["source"])
+        tag = f"cell_{i}"
+        try:
+            exec(compile(src, tag, "exec"), ns)
+            results.append({"cell": i, "status": "ok"})
+        except Exception as exc:
+            results.append({"cell": i, "status": "error", "error": str(exc)})
+            traceback.print_exc()
+            print(f"Stopping at cell {i} due to error.")
+            break
+
+    results_summary = {
+        "status": "completed",
+        "total": len(cells_to_run),
+        "cells": results,
+        "outputs": {
+            "data_dir": str(paths.data_dir),
+            "output_dir": str(paths.output_dir),
+        },
+    }
+    return results_summary
+
+
+if __name__ == "__main__":
+    execute_notebook()
@@ -1,32 +1,52 @@
-"""
-Part 2: 运行完整的 notebook cells 1-35
-解决中文路径编码问题
-"""
-import warnings, time, os, sys, json, traceback
-warnings.filterwarnings('ignore')
+import warnings
+
+warnings.filterwarnings("ignore")
+
 import matplotlib
-matplotlib.use('Agg')
-import matplotlib.pyplot as _p
-_p.show = lambda *a, **kw: None

-nb = r'D:\Code\doing_exercises\programs\外教作业外快\强化学习个人课程作业报告\notebooks\insurance_premium_risk.ipynb'
-cells = json.load(open(nb, encoding='utf-8'))['cells']
-code_cells = [c for c in cells if c['cell_type'] == 'code']
-print(f"Total code cells: {len(code_cells)}")
+matplotlib.use("Agg")
+import matplotlib.pyplot as _real_mpl_plt

-main_ns = globals().copy()
-main_ns['RANDOM_STATE'] = 42
+_real_mpl_plt.show = lambda *a, **kw: None

-for i, cell in enumerate(code_cells, start=1):
-    src = ''.join(cell['source'])
-    print(f"\n{'='*60}")
-    print(f"Running cell {i}/{len(code_cells)}...")
-    try:
-        exec(compile(src, f'cell_{i}', 'exec'), main_ns)
-    except Exception as e:
-        print(f"ERROR cell {i}: {e}")
-        traceback.print_exc()
-        print("Stopping.")
-        break
+import os
+import sys
+import time
+import json
+import traceback
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+import seaborn as sns
+from sklearn.metrics import accuracy_score, f1_score, classification_report, confusion_matrix, ConfusionMatrixDisplay
+from sklearn.model_selection import cross_val_score
+from sklearn.preprocessing import StandardScaler, LabelEncoder
+from sklearn.linear_model import LogisticRegression
+from sklearn.ensemble import RandomForestClassifier, VotingClassifier
+from sklearn.pipeline import Pipeline
+from sklearn.compose import ColumnTransformer
+from sklearn.preprocessing import OneHotEncoder
+from sklearn.impute import SimpleImputer
+from sklearn.cluster import KMeans
+from sklearn.mixture import GaussianMixture
+from sklearn.metrics import silhouette_score
+from sklearn.decomposition import PCA
+import xgboost as xgb
+import optuna
+optuna.logging.set_verbosity(optuna.logging.WARNING)

-print("\n\nAll cells executed!")
+from src.notebook_runner import execute_notebook
+from src.runtime_paths import build_paths
+
+paths = build_paths()
+print(f"Project root : {paths.project_root}")
+print(f"Notebook     : {paths.notebook}")
+print(f"Data dir     : {paths.data_dir}")
+print(f"Output dir   : {paths.output_dir}")
+
+ns = vars()
+
+result = execute_notebook(start_at=1, namespace=ns)
+print(f"\nExecution finished: {result['status']}")
+print(f"Cells run: {len([c for c in result['cells'] if c['status'] == 'ok'])}/{result['total']}")
+print(f"Output dir: {result['outputs']['output_dir']}")
@@ -0,0 +1,31 @@
+from dataclasses import dataclass
+from pathlib import Path
+
+
+@dataclass(frozen=True)
+class RuntimePaths:
+    project_root: Path
+    notebook: Path
+    data_dir: Path
+    output_dir: Path
+
+    def ensure_outputs(self) -> None:
+        (self.output_dir / "figures").mkdir(parents=True, exist_ok=True)
+        (self.output_dir / "tables").mkdir(parents=True, exist_ok=True)
+        (self.output_dir / "predictions").mkdir(parents=True, exist_ok=True)
+
+    def as_injection(self) -> dict:
+        return {
+            "DATA_DIR": str(self.data_dir),
+            "OUTPUT_DIR": str(self.output_dir),
+        }
+
+
+def build_paths() -> RuntimePaths:
+    root = Path(__file__).resolve().parents[1]
+    return RuntimePaths(
+        project_root=root,
+        notebook=root / "notebooks" / "insurance_premium_risk.ipynb",
+        data_dir=root / "dataset_final",
+        output_dir=root / "outputs",
+    )