generated from LufsX/Experiment-LogReg
This commit is contained in:
+24
-46
@@ -26,47 +26,6 @@ name = "邓龙树"
|
||||
|
||||
import numpy as np
|
||||
|
||||
def sigmoid(z):
|
||||
"""计算sigmoid函数"""
|
||||
return 1 / (1 + np.exp(-z))
|
||||
|
||||
def train_logreg(X, y, learning_rate=0.01, 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。
|
||||
- learning_rate (float): 学习率,默认值为0.01。
|
||||
- iterations (int): 迭代次数,默认值为1000。
|
||||
|
||||
Returns:
|
||||
- numpy array: 训练得到的权重。
|
||||
"""
|
||||
m, n = X.shape
|
||||
weights = np.zeros(n) # 初始化权重
|
||||
for _ in range(iterations):
|
||||
z = np.dot(X, weights)
|
||||
predictions = sigmoid(z)
|
||||
error = predictions - y
|
||||
gradient = np.dot(X.T, error) / m
|
||||
weights -= learning_rate * gradient # 更新权重
|
||||
return weights
|
||||
|
||||
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。
|
||||
"""
|
||||
probability = sigmoid(np.dot(inX, weights))
|
||||
return 1 if probability >= 0.5 else 0 # 二分类阈值为0.5
|
||||
|
||||
def load_data(filepath):
|
||||
"""
|
||||
从文件中加载数据。
|
||||
@@ -78,17 +37,36 @@ def load_data(filepath):
|
||||
- np.array: 特征数据。
|
||||
- np.array: 标签数据。
|
||||
"""
|
||||
# 使用dtype=str先将所有数据加载为字符串
|
||||
data = np.loadtxt(filepath, delimiter=' ', dtype=str)
|
||||
# 读取每一行作为字符串
|
||||
with open(filepath, 'r') as file:
|
||||
lines = file.readlines()
|
||||
|
||||
# 替换'?'为np.nan,并转换为浮点数
|
||||
data[data == '?'] = np.nan
|
||||
data = data.astype(float)
|
||||
# 分割每一行并填充缺失的列
|
||||
max_cols = 0
|
||||
rows = []
|
||||
for line in lines:
|
||||
# 分割每一行
|
||||
cols = line.strip().split(' ')
|
||||
# 记录最大列数
|
||||
max_cols = max(max_cols, len(cols))
|
||||
rows.append(cols)
|
||||
|
||||
# 将所有行填充到相同的列数
|
||||
for row in rows:
|
||||
while len(row) < max_cols:
|
||||
row.append('nan')
|
||||
|
||||
# 转换为numpy数组
|
||||
data = np.array(rows, dtype=float)
|
||||
|
||||
# 用np.nan替换字符串'nan'
|
||||
data[data == 'nan'] = np.nan
|
||||
|
||||
X = data[:, :-1] # 假设最后一列是标签
|
||||
y = data[:, -1]
|
||||
return X, y
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
# 加载训练集
|
||||
X_train, y_train = load_data("./data/horse-colic.data")
|
||||
|
||||
Reference in New Issue
Block a user