generated from LufsX/Experiment-LogReg
This commit is contained in:
+8
-12
@@ -29,15 +29,15 @@ def sigmoid(z):
|
||||
|
||||
|
||||
# 对率回归训练函数
|
||||
def train_logreg(X, y, learning_rate=0.01, iterations=1000):
|
||||
def train_logreg(X, y, learning_rate=0.05, iterations=500):
|
||||
"""
|
||||
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。
|
||||
- learning_rate (float): 学习率, 默认值为0.05。
|
||||
- iterations (int): 迭代次数, 默认值为500。
|
||||
|
||||
Returns:
|
||||
- numpy array: 训练得到的权重和偏置(合并为一个数组)。
|
||||
@@ -49,7 +49,7 @@ def train_logreg(X, y, learning_rate=0.01, iterations=1000):
|
||||
X = np.c_[X, np.ones((n_samples, 1))]
|
||||
|
||||
# 梯度下降法
|
||||
for i in range(iterations):
|
||||
for _ in range(iterations):
|
||||
linear_model = np.dot(X, weights)
|
||||
y_pred = sigmoid(linear_model)
|
||||
|
||||
@@ -84,7 +84,7 @@ def predict_logreg(inX, weights):
|
||||
# 数据加载函数
|
||||
def load_data(filepath):
|
||||
"""
|
||||
从文件中加载数据,处理缺失值,将 ? 替换为 NaN 并填充缺失值。
|
||||
从文件中加载数据,处理缺失值,将 ? 替换为 NaN 并填充缺失值为 0。
|
||||
|
||||
Args:
|
||||
- filepath (str): 数据文件路径。
|
||||
@@ -94,12 +94,7 @@ def load_data(filepath):
|
||||
- np.array: 标签数据。
|
||||
"""
|
||||
# 使用 genfromtxt 读取文件并将缺失值标记为 NaN
|
||||
data = np.genfromtxt(filepath, delimiter=' ', dtype=float, missing_values='?', filling_values=np.nan)
|
||||
|
||||
# 对每一列的 NaN 值用列的平均值进行填充
|
||||
nan_columns = np.isnan(data)
|
||||
mean_values = np.nanmean(data, axis=0)
|
||||
data[nan_columns] = np.take(mean_values, np.where(nan_columns)[1])
|
||||
data = np.genfromtxt(filepath, delimiter=' ', dtype=float, missing_values='?', filling_values=0.0)
|
||||
|
||||
# 分离特征和标签
|
||||
X = data[:, :-1] # 所有列(除了最后一列)是特征
|
||||
@@ -115,7 +110,7 @@ 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=1000)
|
||||
weights = train_logreg(X_train, y_train, learning_rate=0.05, iterations=500)
|
||||
|
||||
# 测试单个数据
|
||||
y_pred = predict_logreg(X_test[11], weights)
|
||||
@@ -123,3 +118,4 @@ if __name__ == "__main__":
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user