diff --git a/Scripts/main_RF.py b/Scripts/main_RF.py
index 79401a50a4a04ec9810d4ef0793e24def2deaf7e..1b58cd3f933899d431cd9dc98189ef008d94909f 100644
--- a/Scripts/main_RF.py
+++ b/Scripts/main_RF.py
@@ -8,13 +8,13 @@ from sklearn.ensemble import RandomForestClassifier
 from sklearn.model_selection import train_test_split
 from sklearn.metrics import classification_report, confusion_matrix
 
-# === 1. 读取数据 ===
+
 df = pd.read_csv("../Dataset/Linhac24-25_Sportlogiq.csv", encoding="utf-8")
 
-# === 2. 创建目标变量 ===
+
 df['is_goal'] = (df['eventname'] == 'goal').astype(int)
 
-# === 3. 选择特征 ===
+#选择特征
 features = [
     'xg_allattempts', 'compiledgametime', 'period', 'scoredifferential',
     'teamskatersonicecount', 'opposingteamskatersonicecount',
@@ -25,17 +25,17 @@ features = [
 
 df = df[features + ['is_goal']].copy()
 
-# === 4. 数据预处理 ===
+
 df.fillna({'xg_allattempts': 0, 'compiledgametime': 0, 'scoredifferential': 0,
            'teamskatersonicecount': 5, 'opposingteamskatersonicecount': 5,
            'xadjcoord': 0, 'yadjcoord': 0}, inplace=True)
 
-# 编码分类变量
+
 categorical_cols = ['teaminpossession', 'currentpossession', 'manpowersituation',
                     'ishomegame', 'playerprimaryposition', 'type', 'outcome']
 df = pd.get_dummies(df, columns=categorical_cols, drop_first=True)
 
-# === 5. 拆分数据集 ===
+#拆分数据集
 X = df.drop(columns=['is_goal'])
 y = df['is_goal']
 
@@ -43,23 +43,23 @@ X_train, X_test, y_train, y_test = train_test_split(
     X, y, test_size=0.2, stratify=y, random_state=42
 )
 
-# === 6. 模型路径设置 ===
+#模型路径
 model_dir = os.path.join(os.path.dirname(__file__), "../Model")
 os.makedirs(model_dir, exist_ok=True)
 model_path = os.path.join(model_dir, "RF1.pkl")
 
-# === 7. 加载或训练模型 ===
+
 if os.path.exists(model_path):
-    print("✔ 模型已存在，正在加载模型...")
+    print("模型已存在，正在加载模型...")
     rf = joblib.load(model_path)
 else:
-    print("⚙️ 正在训练模型...")
+    print("正在训练模型...")
     rf = RandomForestClassifier(n_estimators=100, random_state=42)
     rf.fit(X_train, y_train)
     joblib.dump(rf, model_path)
-    print(f"✅ 模型已保存到: {model_path}")
+    print(f"模型已保存到: {model_path}")
 
-# === 8. 模型评估 ===
+#模型评估
 y_pred = rf.predict(X_test)
 
 print("\n=== Classification Report ===")
@@ -68,7 +68,6 @@ print(classification_report(y_test, y_pred))
 print("=== Confusion Matrix ===")
 print(confusion_matrix(y_test, y_pred))
 
-# === 9. 特征重要性可视化 ===
 importances = pd.Series(rf.feature_importances_, index=X.columns)
 top_importances = importances.sort_values(ascending=False).head(20)
 
@@ -79,9 +78,8 @@ plt.title("Top 20 Important Features for Predicting Goals")
 plt.xlabel("Feature Importance")
 plt.tight_layout()
 
-# 👉 保存图表到指定位置（例如模型文件夹）
 plot_path = os.path.join("../Graph", "RF1.jpg")
 plt.savefig(plot_path, dpi=300, bbox_inches='tight')
-print(f"📊 图表已保存到: {plot_path}")
+print(f"图表已保存到: {plot_path}")
 
 plt.show()
\ No newline at end of file
diff --git a/Scripts/main_RF2.py b/Scripts/main_RF2.py
index 29fd0452f3ed75f67f08f8f225622528b54791ac..a15f04991ce1808afb4cdab7750b367a636b2c1e 100644
--- a/Scripts/main_RF2.py
+++ b/Scripts/main_RF2.py
@@ -6,43 +6,37 @@ from sklearn.metrics import classification_report, confusion_matrix
 from sklearn.preprocessing import LabelEncoder
 import os
 
-# 1. 读取数据
-df = pd.read_csv("../Dataset/Linhac24-25_Sportlogiq.csv", encoding="utf-8")  # 修改为你的文件名和分隔符
 
-# 2. 目标列：是否为进球
+df = pd.read_csv("../Dataset/Linhac24-25_Sportlogiq.csv", encoding="utf-8")
+
 df['is_goal'] = (df['eventname'] == 'goal').astype(int)
 
-# 3. 删除无用或无关紧要的列（如gameid、playerid等）
+#删除无用或无关紧要的列（如gameid、playerid等）
 drop_cols = ['gameid', 'playerid', 'eventname', 'compiledgametime']
 df.drop(columns=drop_cols, inplace=True, errors='ignore')
 
-# 4. 编码分类变量
 categorical_cols = df.select_dtypes(include='object').columns
 for col in categorical_cols:
     df[col] = LabelEncoder().fit_transform(df[col].astype(str))
 
-# 5. 分离特征与标签
 X = df.drop(columns=['is_goal'])
 y = df['is_goal']
 
-# 6. 分割训练集与测试集
+
 X_train, X_test, y_train, y_test = train_test_split(
     X, y, test_size=0.2, stratify=y, random_state=42
 )
 
 
-
-
-# 7. 建立模型，使用 class_weight 处理不平衡
+# 使用 class_weight 处理不平衡
 model = RandomForestClassifier(
     n_estimators=100,
     random_state=42,
-    class_weight='balanced',  # 核心修改
+    class_weight='balanced',
     n_jobs=-1
 )
 model.fit(X_train, y_train)
 
-#保存模型
 joblib.dump(model, "../Model/RF2.pkl")
 print("模型已保存")
 
@@ -51,7 +45,6 @@ print("模型已保存")
 
 model = joblib.load("../Model/RF2.pkl")
 
-# 8. 预测与评估
 y_pred = model.predict(X_test)
 
 print("=== Classification Report ===")
@@ -60,7 +53,6 @@ print(classification_report(y_test, y_pred, digits=4))
 print("\n=== Confusion Matrix ===")
 print(confusion_matrix(y_test, y_pred))
 
-# 9. 特征重要性（可选）
 importances = pd.Series(model.feature_importances_, index=X.columns)
 print("\n=== Top 10 Important Features ===")
 print(importances.sort_values(ascending=False).head(10))
diff --git a/Scripts/main_RF3_visual.py b/Scripts/main_RF3_visual.py
index b61b56f6577410d7ac3192e401b51c6c4323bb03..61544b9eb67bbb67e47a7fc9587ddef968356a6c 100644
--- a/Scripts/main_RF3_visual.py
+++ b/Scripts/main_RF3_visual.py
@@ -7,31 +7,25 @@ from sklearn.ensemble import RandomForestClassifier
 from sklearn.metrics import classification_report, confusion_matrix
 from sklearn.preprocessing import LabelEncoder
 
-# 1. 读取数据
+
 df = pd.read_csv("../Dataset/Linhac24-25_Sportlogiq.csv", encoding="utf-8")
 
-# 2. 设置目标列：是否为进球
 df['is_goal'] = (df['eventname'] == 'goal').astype(int)
 
-# 3. 删除无用列
 drop_cols = ['gameid', 'playerid', 'eventname', 'compiledgametime']
 df.drop(columns=drop_cols, inplace=True, errors='ignore')
 
-# 4. 编码分类变量
 categorical_cols = df.select_dtypes(include='object').columns
 for col in categorical_cols:
     df[col] = LabelEncoder().fit_transform(df[col].astype(str))
 
-# 5. 分离特征与标签
 X = df.drop(columns=['is_goal'])
 y = df['is_goal']
 
-# 6. 划分训练集与测试集
 X_train, X_test, y_train, y_test = train_test_split(
     X, y, test_size=0.2, stratify=y, random_state=42
 )
 
-# 7. 建立并训练随机森林模型（处理类别不平衡）
 model = RandomForestClassifier(
     n_estimators=100,
     random_state=42,
@@ -40,7 +34,6 @@ model = RandomForestClassifier(
 )
 model.fit(X_train, y_train)
 
-# 9. 加载模型并进行评估
 model = joblib.load('../Model/RF2.pkl')
 y_pred = model.predict(X_test)
 
@@ -50,17 +43,15 @@ print(classification_report(y_test, y_pred, digits=4))
 print("\n=== Confusion Matrix ===")
 print(confusion_matrix(y_test, y_pred))
 
-# 10. 特征重要性
+# 特征重要性
 importances = pd.Series(model.feature_importances_, index=X.columns)
 print("\n=== Top 10 Important Features ===")
 print(importances.sort_values(ascending=False).head(10))
 
-# 11. 可视化进球位置
 
 goal_data = df[df['is_goal'] == 1]
 
-# ——【方式一】KDE 热力图（平滑）——
-# ——【方式一】KDE 热力图（平滑）——
+#KDE 热力图
 plt.figure(figsize=(10, 6))
 sns.kdeplot(
     x=goal_data['xadjcoord'],
@@ -79,14 +70,13 @@ plt.ylabel('Y Coordinate')
 plt.grid(True)
 plt.legend()
 
-# ✅ 保存 KDE 图像
 plt.tight_layout()
 plt.savefig("../Graph/goal_heatmap_KDE.png", dpi=300, bbox_inches='tight')
-print("✅ KDE 热力图已保存为: ../Model/goal_heatmap_KDE.png")
+print("KDE 热力图已保存为: ../Model/goal_heatmap_KDE.png")
 
 plt.show()
 
-# ——【方式二】Hexbin 热力图（像素块）——
+# Hexbin 热力图
 plt.figure(figsize=(10, 6))
 plt.hexbin(
     goal_data['xadjcoord'],
@@ -105,9 +95,8 @@ plt.ylabel('Y Coordinate')
 plt.grid(True)
 plt.legend()
 
-# ✅ 保存 Hexbin 图像
 plt.tight_layout()
 plt.savefig("../Graph/goal_heatmap_Hexbin.png", dpi=300, bbox_inches='tight')
-print("✅ Hexbin 热力图已保存为: ../Model/goal_heatmap_Hexbin.png")
+print("Hexbin 热力图已保存为: ../Model/goal_heatmap_Hexbin.png")
 
 plt.show()
diff --git a/Scripts/main_XGB.py b/Scripts/main_XGB.py
index e264e59ee7d26ab8ef258cbe9af5870fb3ba7441..824cf6b41c353dbd16d35836b286cffa27757f49 100644
--- a/Scripts/main_XGB.py
+++ b/Scripts/main_XGB.py
@@ -7,16 +7,13 @@ import joblib
 from sklearn.model_selection import train_test_split
 from sklearn.metrics import classification_report, confusion_matrix
 
-plt.rcParams['font.family'] = 'Microsoft YaHei'  # 设置中文字体为微软雅黑
-plt.rcParams['axes.unicode_minus'] = False       # 正确显示负号
+plt.rcParams['font.family'] = 'Microsoft YaHei'
+plt.rcParams['axes.unicode_minus'] = False
 
-# === 1. 读取数据 ===
 df = pd.read_csv("../Dataset/Linhac24-25_Sportlogiq.csv")
 
-# === 2. 构造目标变量 ===
 df["is_goal"] = (df["eventname"] == "goal").astype(int)
 
-# === 3. 选择特征 ===
 feature_cols = [
     "xadjcoord", "yadjcoord", "compiledgametime",
     "scoredifferential", "teamskatersonicecount", "opposingteamskatersonicecount",
@@ -24,37 +21,32 @@ feature_cols = [
     "type", "manpowersituation", "playerprimaryposition", "outcome"
 ]
 
-# === 4. 预处理 ===
 df["xg_allattempts"] = df["xg_allattempts"].fillna(0)
 df["compiledgametime"] = df["compiledgametime"].fillna(0)
 
 # One-hot 编码
 df_encoded = pd.get_dummies(df[feature_cols], drop_first=True)
 
-# 特征 & 标签
+#特征
 X = df_encoded
 y = df["is_goal"]
 
-# === 5. 划分训练和测试集 ===
 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
 
-# === 6. 模型保存路径 ===
 model_dir = os.path.join(os.path.dirname(__file__), "../Model")
 os.makedirs(model_dir, exist_ok=True)
 model_path = os.path.join(model_dir, "XGB.pkl")
 
-# === 7. 加载或训练模型 ===
 if os.path.exists(model_path):
-    print("✔ 已检测到已有模型，正在加载中...")
+    print("已检测到已有模型，正在加载中...")
     model = joblib.load(model_path)
 else:
-    print("⚙️ 正在训练 XGBoost 模型...")
+    print("正在训练 XGBoost 模型...")
     model = xgb.XGBClassifier(eval_metric='logloss', use_label_encoder=False)
     model.fit(X_train, y_train)
     joblib.dump(model, model_path)
-    print(f"✅ 模型已保存到: {model_path}")
+    print(f"模型已保存到: {model_path}")
 
-# === 8. 模型评估 ===
 y_pred = model.predict(X_test)
 print("\n=== 分类报告 ===")
 print(classification_report(y_test, y_pred))
diff --git a/Scripts/main_XGB2.py b/Scripts/main_XGB2.py
index 5f3d0e5c27e675cedefdd6a6094c75826b1e80bf..4c94e417482f4c7d0b5fd95b11af47431fbde028 100644
--- a/Scripts/main_XGB2.py
+++ b/Scripts/main_XGB2.py
@@ -6,17 +6,14 @@ import matplotlib.pyplot as plt
 from sklearn.model_selection import train_test_split
 from sklearn.metrics import classification_report, confusion_matrix
 
-# 设置中文字体
 plt.rcParams['font.family'] = 'Microsoft YaHei'
 plt.rcParams['axes.unicode_minus'] = False
 
-# 1. 读取数据
 df = pd.read_csv("../Dataset/Linhac24-25_Sportlogiq.csv")
 
-# 2. 构造目标变量
 df["is_goal"] = (df["eventname"] == "goal").astype(int)
 
-# 3. 选择特征列
+
 feature_cols = [
     "xadjcoord", "yadjcoord", "compiledgametime",
     "scoredifferential", "teamskatersonicecount", "opposingteamskatersonicecount",
@@ -24,41 +21,33 @@ feature_cols = [
     "type", "manpowersituation", "playerprimaryposition", "outcome"
 ]
 
-# 4. 预处理
 df["xg_allattempts"] = df["xg_allattempts"].fillna(0)
 df["compiledgametime"] = df["compiledgametime"].fillna(0)
 
-# 类别变量 one-hot 编码
 df_encoded = pd.get_dummies(df[feature_cols], drop_first=True)
 
-# 特征和标签
 X = df_encoded
 y = df["is_goal"]
 
-# 5. 划分训练/测试集
 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
 
-# 6. 训练模型
 model = xgb.XGBClassifier(eval_metric='logloss')
 model.fit(X_train, y_train)
 
-# ✅ 保存模型
 #model.save_model("../Model/XGB2.pkl")
 
-# ✅ 加载模型
 loaded_model = xgb.XGBClassifier()
 loaded_model.load_model("../Model/XGB2.pkl")
 
-# 7. 模型评估
 y_pred = loaded_model.predict(X_test)
 print(classification_report(y_test, y_pred))
 print("混淆矩阵：")
 print(confusion_matrix(y_test, y_pred))
 
-# 8. 获取预测概率
+
 y_prob = loaded_model.predict_proba(X_test)[:, 1]
 
-# 9. 组合 df_test 用于分析
+#组合 df_test 用于分析
 df_test = X_test.copy()
 df_test['is_goal'] = y_test
 df_test['goal_prob'] = y_prob
@@ -67,7 +56,7 @@ df_test['yadjcoord'] = df.loc[df_test.index, 'yadjcoord']
 df_test['manpowersituation'] = df.loc[df_test.index, 'manpowersituation']
 df_test['scoredifferential'] = df.loc[df_test.index, 'scoredifferential']
 
-# 10. 进球位置与进球概率关系图
+#进球位置与进球概率关系图
 plt.figure(figsize=(10, 6))
 plt.scatter(df_test['xadjcoord'], df_test['yadjcoord'], c=df_test['goal_prob'], cmap='coolwarm', alpha=0.5)
 plt.colorbar(label="Goal Probability")
@@ -82,7 +71,7 @@ plt.show()
 
 
 
-# 11. 局势 vs 进球概率
+#局势 vs 进球概率
 plt.figure(figsize=(12, 6))
 sns.boxplot(x=df_test['manpowersituation'], y=df_test['goal_prob'])
 plt.title("Goal Probability Across Different Manpower Situations")
@@ -94,7 +83,7 @@ plt.savefig("../Graph/goal_prob_by_manpowersituation.png", dpi=300, bbox_inches=
 plt.show()
 
 
-# 12. 比分差 vs 进球概率
+# 比分差 vs 进球概率
 plt.figure(figsize=(10, 6))
 sns.lineplot(x=df_test['scoredifferential'], y=df_test['goal_prob'], marker='o')
 plt.title("Goal Probability vs Score Differential")
diff --git a/Scripts/process_1.py b/Scripts/process_1.py
index 84c1e8de038cbb27de6b38bf14a0f8c9c90609a4..840e7b71591406da3c7d48c8538015053b7e70a5 100644
--- a/Scripts/process_1.py
+++ b/Scripts/process_1.py
@@ -1,38 +1,35 @@
 import pandas as pd
 import matplotlib.pyplot as plt
 import seaborn as sns
-plt.rcParams['font.family'] = 'Microsoft YaHei'  # 设置中文字体为微软雅黑
-plt.rcParams['axes.unicode_minus'] = False       # 正确显示负号
-# 读取数据（假设你已经保存为 CSV 文件）
+plt.rcParams['font.family'] = 'Microsoft YaHei'
+plt.rcParams['axes.unicode_minus'] = False
+
 df = pd.read_csv("Linhac24-25_Sportlogiq.csv")
 
-# 显示基本信息
+
 print(df.info())
 print(df.head())
 
-
 event_counts = df['eventname'].value_counts()
 print("事件类型统计:\n", event_counts)
 
-
 team_possession_counts = df['teaminpossession'].value_counts()
 print("控球队出现次数:\n", team_possession_counts)
 
-
 df['xg_allattempts'] = pd.to_numeric(df['xg_allattempts'], errors='coerce')
 xg_by_team = df.groupby('teamid')['xg_allattempts'].sum()
 print("每支球队的总xG:\n", xg_by_team)
 
-
 player_actions = df['playerid'].value_counts().head(10)
 print("参与最多事件的前10位球员:\n", player_actions)
 
-
-
 success_rate = df['outcome'].value_counts(normalize=True)
 print("事件成功与失败比例:\n", success_rate)
 
 
+
+
+
 print("\n简单总结：")
 print(f"总事件数: {len(df)}")
 print(f"总xG: {df['xg_allattempts'].sum():.2f}")
diff --git a/Scripts/process_2.py b/Scripts/process_2.py
index c5018e18c8967cbf8e777aed901baa9446631f54..c025647c9c7b385305ecf69758f11d22fa63d06e 100644
--- a/Scripts/process_2.py
+++ b/Scripts/process_2.py
@@ -10,12 +10,12 @@ df = pd.read_csv("Linhac24-25_Sportlogiq.csv")
 
 sns.set(style="whitegrid")
 
-print("📌 数据基本信息：")
+print("数据基本信息：")
 print(df.info())
-print("\n📈 描述性统计：")
+print("\n描述性统计：")
 print(df.describe())
 
-print("\n📊 不同事件类型分布：")
+print("\n不同事件类型分布：")
 print(df['eventname'].value_counts())
 
 plt.figure(figsize=(12, 5))