generated from LufsX/Experiment-LogReg
This commit is contained in:
+42
-39
@@ -1,55 +1,42 @@
|
|||||||
# function.py
|
|
||||||
import numpy as np
|
import numpy as np
|
||||||
|
|
||||||
"""
|
"""
|
||||||
实验说明:
|
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 = "202210423218"
|
id = "202210423218"
|
||||||
name = "唐一鸣"
|
name = "唐一鸣"
|
||||||
|
|
||||||
|
def sigmoid(z):
|
||||||
|
return 1 / (1 + np.exp(-z))
|
||||||
|
|
||||||
# 这里写自己的代码实现
|
def train_logreg(X, y, other_params=""):
|
||||||
def train_logreg(X, y, other_params="你自己设置一些其它参数"):
|
|
||||||
"""
|
"""
|
||||||
对率回归训练函数。
|
对率回归训练函数。
|
||||||
|
|
||||||
Args:
|
Args:
|
||||||
- X (numpy array): 特征数据。
|
- X (numpy array): 特征数据。
|
||||||
- y (numpy array): 标签数据。
|
- y (numpy array): 标签数据。
|
||||||
- other_params (str): 其它参数,可选设置。若需设置其它参数,必须设置其默认值,否则将**无法通过**自动评测,**成绩为零**。
|
- other_params (str): 其它参数,可选设置。
|
||||||
|
|
||||||
Returns:
|
Returns:
|
||||||
- numpy array: 训练得到的权重。
|
- numpy array: 训练得到的权重。
|
||||||
"""
|
"""
|
||||||
m, n = X.shape # 样本数和特征数
|
m, n = X.shape
|
||||||
weights = np.zeros(n) # 初始化权重
|
weights = np.zeros(n)
|
||||||
for _ in range(iterations):
|
learning_rate = 0.01
|
||||||
# 计算当前模型的预测值
|
num_iterations = 1000
|
||||||
z = np.dot(X, weights)
|
|
||||||
h = sigmoid(z)
|
|
||||||
# 计算损失函数的梯度
|
|
||||||
gradient = np.dot(X.T, (h - y)) / m
|
|
||||||
# 更新权重
|
|
||||||
weights -= learning_rate * gradient
|
|
||||||
return weights
|
|
||||||
|
|
||||||
|
for _ in range(num_iterations):
|
||||||
|
linear_model = np.dot(X, weights)
|
||||||
|
y_predicted = sigmoid(linear_model)
|
||||||
|
|
||||||
|
# Gradient descent
|
||||||
|
gradient = np.dot(X.T, (y_predicted - y)) / m
|
||||||
|
weights -= learning_rate * gradient
|
||||||
|
|
||||||
|
return weights
|
||||||
|
|
||||||
def predict_logreg(inX, weights):
|
def predict_logreg(inX, weights):
|
||||||
"""
|
"""
|
||||||
@@ -62,24 +49,39 @@ def predict_logreg(inX, weights):
|
|||||||
Returns:
|
Returns:
|
||||||
- int: 预测的分类标签。
|
- int: 预测的分类标签。
|
||||||
"""
|
"""
|
||||||
probability = sigmoid(np.dot(inX, weights))
|
prob = sigmoid(np.dot(inX, weights))
|
||||||
return 1 if probability >= 0.5 else 0
|
return 1 if prob >= 0.5 else 0
|
||||||
|
|
||||||
# 这里也写自己的代码实现
|
|
||||||
def load_data(filepath):
|
def load_data(filepath):
|
||||||
"""
|
"""
|
||||||
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**。
|
从文件中加载数据,处理缺失值。
|
||||||
|
|
||||||
Args:
|
Args:
|
||||||
- directory (str): 数据文件路径。
|
- filepath (str): 数据文件路径。
|
||||||
|
|
||||||
Returns:
|
Returns:
|
||||||
- np.array: 特征数据。
|
- np.array: 特征数据。
|
||||||
- np.array: 标签数据。
|
- np.array: 标签数据。
|
||||||
"""
|
"""
|
||||||
data = np.loadtxt(filepath, delimiter=' ', dtype=float)
|
with open(filepath, 'r') as f:
|
||||||
X = data[:, :-1] # 特征数据
|
lines = f.readlines()
|
||||||
y = data[:, -1] # 标签数据
|
|
||||||
|
data = []
|
||||||
|
for line in lines:
|
||||||
|
# 替换缺失值 '?' 为 np.nan
|
||||||
|
row = line.strip().split()
|
||||||
|
row = [float(x) if x != '?' else np.nan for x in row]
|
||||||
|
data.append(row)
|
||||||
|
|
||||||
|
data = np.array(data)
|
||||||
|
|
||||||
|
# 处理缺失值:用均值替代缺失值
|
||||||
|
data = np.where(np.isnan(data), np.nanmean(data, axis=0), data)
|
||||||
|
|
||||||
|
# 特征为前n-1列,标签为最后一列
|
||||||
|
X = data[:, :-1]
|
||||||
|
y = data[:, -1]
|
||||||
|
|
||||||
return X, y
|
return X, y
|
||||||
|
|
||||||
|
|
||||||
@@ -92,8 +94,9 @@ if __name__ == "__main__":
|
|||||||
X_test, y_test = load_data("./data/horse-colic.test")
|
X_test, y_test = load_data("./data/horse-colic.test")
|
||||||
|
|
||||||
# 训练对率回归模型
|
# 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)
|
||||||
|
|
||||||
# 测试单个数据
|
# 测试单个数据
|
||||||
y_pred = predict_logreg(X_test[11], weights)
|
y_pred = predict_logreg(X_test[11], weights)
|
||||||
print(f"预测结果: {y_pred}, 实际结果: {y_test[11]}")
|
print(f"预测结果: {y_pred}, 实际结果: {y_test[11]}")
|
||||||
|
|
||||||
|
|||||||
Reference in New Issue
Block a user