generated from LufsX/Experiment-LogReg
This commit is contained in:
+88
-84
@@ -1,92 +1,96 @@
|
||||
import numpy as np
|
||||
import pandas as pd
|
||||
|
||||
# 填写学号和姓名
|
||||
id = "20221042323"
|
||||
name = "李顺潮"
|
||||
|
||||
def train_logreg(X, y, other_params="你自己设置一些其它参数", learning_rate=0.01, num_iterations=1000):
|
||||
"""
|
||||
对率回归训练函数。
|
||||
|
||||
Args:
|
||||
- X (numpy array): 特征数据。
|
||||
- y (numpy array): 标签数据。
|
||||
- other_params (str): 其它参数，这里暂未使用。
|
||||
- learning_rate (float): 学习率。
|
||||
- num_iterations (int): 迭代次数。
|
||||
|
||||
Returns:
|
||||
- numpy array: 训练得到的权重。
|
||||
"""
|
||||
# 添加偏置项
|
||||
X_b = np.c_[np.ones((X.shape[0], 1)), X] # 添加一列1作为偏置项
|
||||
m, n = X_b.shape
|
||||
# function.py
|
||||
import numpy as np
|
||||
from sklearn.preprocessing import StandardScaler
|
||||
from sklearn.metrics import accuracy_score
|
||||
"""
|
||||
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，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。
|
||||
|
||||
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
|
||||
"""
|
||||
|
||||
|
||||
"""
|
||||
请在这里填写学号和姓名，不填写即成绩无效
|
||||
"""
|
||||
id = "202210423223"
|
||||
name = "李顺潮"
|
||||
|
||||
|
||||
# 这里写自己的代码实现
|
||||
def train_logreg(X, y, learning_rate=0.01, iterations=1000, lambda_reg=0.1):
|
||||
m, n = X.shape
|
||||
weights = np.zeros(n)
|
||||
|
||||
# 梯度下降
|
||||
for iteration in range(num_iterations):
|
||||
y_pred_prob = 1 / (1 + np.exp(-np.dot(X_b, weights)))
|
||||
dw = (1 / m) * np.dot(X_b.T, (y_pred_prob - y))
|
||||
bias = 0
|
||||
|
||||
for _ in range(iterations):
|
||||
linear_output = np.dot(X, weights) + bias
|
||||
y_predicted = 1 / (1 + np.exp(-linear_output))
|
||||
|
||||
# 计算梯度（包括正则化项）
|
||||
dw = (1 / m) * np.dot(X.T, (y_predicted - y)) + (lambda_reg / m) * weights
|
||||
db = (1 / m) * np.sum(y_predicted - y)
|
||||
|
||||
# 更新权重和偏置
|
||||
weights -= learning_rate * dw
|
||||
bias -= learning_rate * db
|
||||
|
||||
return weights
|
||||
return weights, bias
|
||||
|
||||
def predict_logreg(inX, weights, bias):
|
||||
linear_output = np.dot(inX, weights) + bias
|
||||
y_predicted = 1 if linear_output >= 0 else 0 # 简化阈值判断，因为已经通过sigmoid函数输出
|
||||
# 注意：这里为了与原始代码保持一致，使用了0.5作为阈值，但实际上由于sigmoid输出范围在(0,1)，可以直接比较linear_output和0
|
||||
# return 1 if y_predicted >= 0.5 else 0 # 等价于上面的判断，但多余计算了sigmoid
|
||||
return y_predicted
|
||||
|
||||
# 这里也写自己的代码实现
|
||||
def load_data(filepath):
|
||||
data = []
|
||||
with open(filepath, 'r') as file:
|
||||
for line in file:
|
||||
parts = line.strip().split()
|
||||
# 将缺失值'?'替换为np.nan，然后使用0填充（或选择其他适当的填充策略）
|
||||
features = [float(part) if part != '?' else np.nan for part in parts[:-1]]
|
||||
# 填充NaN值（这里选择0，但也可以考虑使用均值、中位数等）
|
||||
features = np.nan_to_num(features, nan=0.0)
|
||||
# 将标签从2转换为1（如果原始标签是1和2的话），这里假设标签已经是0和1
|
||||
label = int(parts[-1]) # 假设标签已经是0和1，无需转换
|
||||
data.append(features + [label])
|
||||
|
||||
def predict_logreg(inX, weights):
|
||||
"""
|
||||
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。
|
||||
"""
|
||||
# 添加偏置项
|
||||
inX_b = np.c_[1, inX]
|
||||
y_pred_prob = 1 / (1 + np.exp(-np.dot(inX_b, weights)))
|
||||
return 1 if y_pred_prob > 0.5 else 0
|
||||
|
||||
def load_data(filepath):
|
||||
"""
|
||||
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。
|
||||
"""
|
||||
# 使用 pandas 加载数据，假设数据使用空格分隔，并且我们想要跳过第一列（可能是标识符）
|
||||
df = pd.read_csv(filepath, delim_whitespace=True, header=None)
|
||||
data_array = np.array(data)
|
||||
X = data_array[:, :-1]
|
||||
y = data_array[:, -1]
|
||||
|
||||
# 假设最后一列是标签，将其转换为 0 和 1
|
||||
df.iloc[:, -1] = df.iloc[:, -1].apply(lambda x: 1 if x == 2 else 0)
|
||||
|
||||
# 将缺失值（如 '?'）替换为 NaN，然后删除包含 NaN 的行（或者选择其他处理方式）
|
||||
df.replace('?', np.nan, inplace=True)
|
||||
df.dropna(inplace=True)
|
||||
|
||||
# 提取特征和标签
|
||||
X = df.iloc[:, :-1].values # 特征
|
||||
y = df.iloc[:, -1].values # 标签
|
||||
|
||||
return X, y
|
||||
|
||||
if __name__ == "__main__":
|
||||
# 加载训练集
|
||||
X_train, y_train = load_data("./data/horse-colic.data")
|
||||
|
||||
# 数据标准化
|
||||
# 特征缩放（标准化）
|
||||
scaler = StandardScaler()
|
||||
X_train_scaled = scaler.fit_transform(X_train)
|
||||
X_scaled = scaler.fit_transform(X)
|
||||
|
||||
# 训练模型
|
||||
weights = train_logreg(X_train_scaled, y_train)
|
||||
return X_scaled, y
|
||||
|
||||
if __name__ == "__main__":
|
||||
# 加载训练集和测试集
|
||||
X_train, y_train = load_data("./data/horse-colic.data")
|
||||
X_test, y_test = load_data("./data/horse-colic.test")
|
||||
|
||||
# 测试预测（这里示例性地选择一个样本进行预测）
|
||||
sample_index = 0 # 选择第一个样本进行预测
|
||||
y_pred = predict_logreg(X_train_scaled[sample_index], weights)
|
||||
print(f"预测结果: {y_pred}, 实际结果: {y_train[sample_index]}")
|
||||
# 训练对率回归模型
|
||||
weights, bias = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000, lambda_reg=0.1)
|
||||
|
||||
# 测试模型并评估性能
|
||||
y_pred = [predict_logreg(x, weights, bias) for x in X_test]
|
||||
accuracy = accuracy_score(y_test, y_pred)
|
||||
print(f"模型准确率: {accuracy}")
|
||||
|
||||
# 打印单个测试样本的预测结果和实际结果（示例）
|
||||
# 注意：索引11可能需要根据实际测试集调整，这里仅作为示例
|
||||
sample_index = 11 # 假设要测试的样本索引是11（注意：索引可能需要根据实际数据调整）
|
||||
print(f"预测结果: {y_pred[sample_index]}, 实际结果: {y_test[sample_index]}")
|
||||
Reference in New Issue
Block a user