generated from LufsX/Experiment-LogReg
This commit is contained in:
+60
-52
@@ -1,83 +1,91 @@
|
||||
# function.py
|
||||
import numpy as np
|
||||
|
||||
# 填写学号和姓名
|
||||
id = "202210423213" # 请替换为实际学号
|
||||
name = "熊浩燃" # 请替换为实际姓名
|
||||
"""
|
||||
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。
|
||||
注意事项:
|
||||
- 在 function.py 中实现对率回归算法。
|
||||
- 可以自行导入其他库,需要在 requirements.txt 添加依赖。
|
||||
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。
|
||||
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
|
||||
"""
|
||||
|
||||
def sigmoid(z):
|
||||
"""计算Sigmoid函数值"""
|
||||
return 1 / (1 + np.exp(-z))
|
||||
id = "202210423213"
|
||||
name = "熊浩燃"
|
||||
|
||||
def train_logreg(X, y, learning_rate=0.01, iterations=5000, reg_lambda=0.1):
|
||||
# 对率回归训练函数
|
||||
def train_logreg(X, y, learning_rate=0.01, iterations=1000):
|
||||
"""
|
||||
对率回归训练函数,带正则化项。
|
||||
|
||||
对率回归训练函数。
|
||||
Args:
|
||||
- X (numpy array): 特征数据。
|
||||
- y (numpy array): 标签数据。
|
||||
- learning_rate (float): 学习率,默认为0.01。
|
||||
- iterations (int): 梯度下降迭代次数,默认为5000。
|
||||
- reg_lambda (float): L2正则化系数,默认为0.1。
|
||||
|
||||
- iterations (int): 迭代次数,默认为1000。
|
||||
Returns:
|
||||
- numpy array: 训练得到的权重。
|
||||
"""
|
||||
# 数据标准化
|
||||
X = (X - X.mean(axis=0)) / X.std(axis=0)
|
||||
|
||||
# 初始化权重
|
||||
weights = np.zeros(X.shape[1])
|
||||
|
||||
# 批量梯度下降
|
||||
|
||||
for i in range(iterations):
|
||||
predictions = sigmoid(np.dot(X, weights))
|
||||
errors = y - predictions
|
||||
# 更新权重,带L2正则化
|
||||
weights += learning_rate * (np.dot(X.T, errors) - reg_lambda * weights)
|
||||
|
||||
# 预测值:通过对率回归模型得到
|
||||
z = np.dot(X, weights)
|
||||
predictions = 1 / (1 + np.exp(-z))
|
||||
|
||||
# 计算梯度并更新权重
|
||||
gradient = np.dot(X.T, (predictions - y)) / y.size
|
||||
weights -= learning_rate * gradient
|
||||
|
||||
# 可选:输出损失,帮助调试
|
||||
if i % 100 == 0:
|
||||
loss = -np.mean(y * np.log(predictions) + (1 - y) * np.log(1 - predictions))
|
||||
print(f"迭代 {i},损失: {loss}")
|
||||
|
||||
return weights
|
||||
|
||||
def predict_logreg(inX, weights, threshold=0.3):
|
||||
# 对率回归预测函数
|
||||
def predict_logreg(inX, weights):
|
||||
"""
|
||||
对率回归预测函数,包含自定义阈值。
|
||||
|
||||
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。
|
||||
- threshold (float): 分类阈值,默认为0.3。
|
||||
|
||||
Returns:
|
||||
- int: 预测的分类标签。
|
||||
"""
|
||||
probability = sigmoid(np.dot(inX, weights))
|
||||
return 1 if probability >= threshold else 0
|
||||
probability = 1 / (1 + np.exp(-np.dot(inX, weights)))
|
||||
return 1 if probability >= 0.5 else 0
|
||||
|
||||
# 数据加载函数
|
||||
def load_data(filepath):
|
||||
"""
|
||||
从文件中加载数据,并填充缺失值。
|
||||
|
||||
从文件中加载数据。
|
||||
Args:
|
||||
- filepath (str): 数据文件路径。
|
||||
|
||||
Returns:
|
||||
- np.array: 特征数据。
|
||||
- np.array: 标签数据。
|
||||
- numpy array: 特征数据。
|
||||
- numpy array: 标签数据。
|
||||
"""
|
||||
data = np.genfromtxt(filepath, delimiter=" ", dtype=float, missing_values="?", filling_values=np.nan)
|
||||
|
||||
# 对每一列单独填充缺失值
|
||||
for i in range(data.shape[1] - 1): # 最后一列是标签列
|
||||
if np.isnan(data[:, i]).any():
|
||||
# 使用中位数填充数值型数据
|
||||
median = np.nanmedian(data[:, i])
|
||||
data[np.isnan(data[:, i]), i] = median
|
||||
|
||||
# 标准化特征数据
|
||||
X = data[:, :-1]
|
||||
y = data[:, -1]
|
||||
X = (X - X.mean(axis=0)) / X.std(axis=0)
|
||||
|
||||
return X, y
|
||||
data = []
|
||||
labels = []
|
||||
with open(filepath, 'r') as file:
|
||||
for line in file:
|
||||
# 去除无效行,并将数据转换为float格式
|
||||
fields = line.strip().split()
|
||||
if '?' not in fields:
|
||||
# 以最后一列作为标签,其余列作为特征
|
||||
features = [float(x) for x in fields[:-1]]
|
||||
label = float(fields[-1])
|
||||
data.append(features)
|
||||
labels.append(label)
|
||||
|
||||
return np.array(data), np.array(labels)
|
||||
|
||||
if __name__ == "__main__":
|
||||
# 加载训练集
|
||||
@@ -87,8 +95,8 @@ if __name__ == "__main__":
|
||||
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, learning_rate=0.01, iterations=5000, reg_lambda=0.1)
|
||||
weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000)
|
||||
|
||||
# 使用自定义阈值测试单个数据
|
||||
y_pred = predict_logreg(X_test[11], weights, threshold=0.3)
|
||||
# 测试单个数据
|
||||
y_pred = predict_logreg(X_test[11], weights)
|
||||
print(f"预测结果: {y_pred}, 实际结果: {y_test[11]}")
|
||||
|
||||
Reference in New Issue
Block a user