From 480a425485351df03205d36e6b3ec659618cd4de Mon Sep 17 00:00:00 2001
From: mogu-king <1547976000@qq.com>
Date: Sun, 3 Nov 2024 05:23:13 +0800
Subject: [PATCH] =?UTF-8?q?=E6=9B=B4=E6=96=B0=20function.py?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 function.py | 115 +++++++++++++++++++++++++++-------------------------
 1 file changed, 59 insertions(+), 56 deletions(-)

diff --git a/function.py b/function.py
index 36ba124..b982418 100644
--- a/function.py
+++ b/function.py
@@ -1,14 +1,21 @@
 import numpy as np
-from sklearn.tree import DecisionTreeClassifier
+from sklearn.ensemble import GradientBoostingClassifier
+from sklearn.preprocessing import StandardScaler
 
 """
 实验说明:
-- 自行编写实现数据加载与预测病马死亡的对率回归算法。
+- 自行编写实现数据加载与预测病马死亡的梯度提升模型。
 
 注意事项:
-- 在 function.py 中实现对率回归算法。
+- 在 function.py 中实现梯度提升模型。
 - 可以自行导入其他库，需要在 requirements.txt 添加依赖，若未在 requirements.txt 中添加使用到的依赖，则无法自动评测，成绩为零。
-- 自动评测脚本仅会加载 train_logreg、predict_logreg 和 load_data 函数，请勿修改函数名。
+- 自动评测脚本仅会加载 train_gb, predict_gb 和 load_data 函数，请勿修改函数名。
+
+测评环境:
+- CPU: Intel Xeon(R) CPU E3-1231 v3 @ 3.40GHz
+- OS: Ubuntu 22.04.4 LTS
+- Python: 3.10.12
+- numpy: 2.1.1
 """
 
 """
@@ -17,33 +24,36 @@ from sklearn.tree import DecisionTreeClassifier
 id = "202210423208"
 name = "刘义天"
 
-# 加载数据并添加非线性特征
+
 def load_data(filepath):
     """
-    从文件中加载数据并生成非线性特征。
+    从文件中加载数据，若不自行正确实现，则无法通过评测，**不记录成绩**
 
     Args:
     - filepath (str): 数据文件路径
 
     Returns:
-    - X (numpy array): 包含非线性特征的特征数据
+    - X (numpy array): 特征数据
     - y (numpy array): 标签数据
     """
-    data = np.genfromtxt(filepath, delimiter=" ", dtype=str)
-    data[data == '?'] = np.nan
-    data = data.astype(float)
-    col_means = np.nanmean(data, axis=0)
-    inds = np.where(np.isnan(data))
-    data[inds] = np.take(col_means, inds[1])
+    data = []
+    with open(filepath, 'r') as file:
+        for line in file:
+            line = line.replace('?', 'nan')
+            data.append([float(x) if x != 'nan' else np.nan for x in line.split()])
+    
+    data = np.array(data)
     X = data[:, :-1]
     y = data[:, -1]
-    square_features = X ** 2
-    log_features = np.log1p(X)
-    X = np.hstack([X, square_features, log_features])
+    
+    col_means = np.nanmean(X, axis=0)
+    inds = np.where(np.isnan(X))
+    X[inds] = np.take(col_means, inds[1])
+    
     return X, y
 
-# 对率回归训练函数
-def train_logreg(X, y, initial_learning_rate=0.1, num_iterations=2000, regularization_strength=0.1):
+
+def train_gb(X, y, learning_rate=0.1, n_estimators=200, max_depth=5):
     """
     对率回归训练函数。
 
@@ -57,50 +67,43 @@ def train_logreg(X, y, initial_learning_rate=0.1, num_iterations=2000, regulariz
     Returns:
     - weights (numpy array): 训练得到的权重。
     """
-   
-    m, n = X.shape
-    weights = np.random.randn(n) * 0.01  # 随机初始化权重
-    learning_rate = initial_learning_rate
-    for i in range(num_iterations):
-        predictions = 1 / (1 + np.exp(-np.dot(X, weights)))
-        gradient = (1 / m) * np.dot(X.T, (predictions - y)) + regularization_strength * weights
-        weights -= learning_rate * gradient
-        if i % 100 == 0:  # 每100次迭代降低学习率
-            learning_rate *= 0.9
-    return weights
-
-# 决策树训练函数
-def train_decision_tree(X, y):
-    clf = DecisionTreeClassifier(max_depth=5, random_state=42)
-    clf.fit(X, y)
-    return clf
-
-# 对率回归预测函数
-def predict_logreg(inX, weights):
-    probability = 1 / (1 + np.exp(-np.dot(inX, weights)))
-    return 1 if probability >= 0.5 else 0
+    scaler = StandardScaler()
+    X = scaler.fit_transform(X)
+    
+    gb_model = GradientBoostingClassifier(
+        learning_rate=learning_rate, 
+        n_estimators=n_estimators, 
+        max_depth=max_depth, 
+        random_state=42
+    )
+    gb_model.fit(X, y)
+    return gb_model, scaler
 
 
-# 融合预测函数
-def predict_fusion(inX, weights, clf, alpha=0.6):
-    prob_logreg = 1 / (1 + np.exp(-np.dot(inX, weights)))
-    prob_tree = clf.predict_proba(inX.reshape(1, -1))[0, 1]
-    final_prob = alpha * prob_logreg + (1 - alpha) * prob_tree
-    return 1 if final_prob >= 0.5 else 0
+def predict_gb(inX, model, scaler):
+    """
+    对率回归预测函数。
+
+    Args:
+    - inX (numpy array): 输入待分类的样本。
+    - weights (numpy array): 训练得到的权重。
+
+    Returns:
+    - int: 预测的分类标签。
+    """
+    inX = scaler.transform([inX])
+    prediction = model.predict(inX)
+    return int(prediction[0])
+
 
 if __name__ == "__main__":
+    # 加载训练和测试数据
     X_train, y_train = load_data("./data/horse-colic.data")
     X_test, y_test = load_data("./data/horse-colic.test")
 
-    # 训练对率回归模型
-    weights = train_logreg(X_train, y_train, initial_learning_rate=0.1, num_iterations=2000)
-
-    # 训练决策树模型
-    clf = train_decision_tree(X_train, y_train)
+    # 训练模型
+    model, scaler = train_gb(X_train, y_train)
 
     # 测试单个数据
-    y_pred_logreg = predict_logreg(X_test[11], weights)
-    y_pred_fusion = predict_fusion(X_test[11], weights, clf)
-
-    print(f"对率回归预测结果: {y_pred_logreg}, 实际结果: {y_test[11]}")
-    print(f"融合预测结果: {y_pred_fusion}, 实际结果: {y_test[11]}")
\ No newline at end of file
+    y_pred = predict_gb(X_test[11], model, scaler)
+    print(f"预测结果: {y_pred}, 实际结果: {int(y_test[11])}")