generated from LufsX/Experiment-LogReg
This commit is contained in:
+23
-45
@@ -10,62 +10,43 @@ from sklearn.linear_model import LogisticRegression
|
|||||||
- 在 function.py 中实现对率回归算法。
|
- 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。
|
- 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。
|
- 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。
|
请在这里填写学号和姓名,不填写即成绩无效。
|
||||||
"""
|
"""
|
||||||
id = "202210423230"
|
id = "202210423xxx"
|
||||||
name = "尹家立"
|
name = "test_name"
|
||||||
|
|
||||||
# Sigmoid function
|
def train_logreg(X, y, max_iter=1000):
|
||||||
def sigmoid(inX):
|
|
||||||
"""Sigmoid激活函数"""
|
|
||||||
return 1.0 / (1 + np.exp(-inX))
|
|
||||||
|
|
||||||
def train_logreg(X, y, learning_rate=0.01, iterations=5000, solver='liblinear'):
|
|
||||||
"""
|
"""
|
||||||
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。
|
||||||
- learning_rate (float): 学习速率。
|
- max_iter (int): 最大迭代次数。
|
||||||
- iterations (int): 最大迭代次数。
|
|
||||||
- solver (str): 使用的求解器。默认值为'liblinear'。
|
|
||||||
|
|
||||||
Returns:
|
Returns:
|
||||||
- numpy array: 训练得到的权重。
|
- LogisticRegression: 训练好的模型。
|
||||||
"""
|
"""
|
||||||
# 使用liblinear作为求解器进行对率回归训练
|
classifier = LogisticRegression(solver='liblinear', max_iter=max_iter)
|
||||||
model = LogisticRegression(solver=solver, max_iter=iterations)
|
classifier.fit(X, y)
|
||||||
model.fit(X, y)
|
return classifier
|
||||||
|
|
||||||
# 返回模型的系数作为权重
|
|
||||||
return model.coef_.flatten()
|
|
||||||
|
|
||||||
def predict_logreg(inX, weights):
|
def predict_logreg(inX, model):
|
||||||
"""
|
"""
|
||||||
对率回归预测函数。
|
对率回归预测函数。
|
||||||
|
|
||||||
Args:
|
Args:
|
||||||
- inX (numpy array): 输入待分类的样本。
|
- inX (numpy array): 输入待分类的样本。
|
||||||
- weights (numpy array): 训练得到的权重。
|
- model (LogisticRegression): 训练好的逻辑回归模型。
|
||||||
|
|
||||||
Returns:
|
Returns:
|
||||||
- int: 预测的分类标签。
|
- int: 预测的分类标签。
|
||||||
"""
|
"""
|
||||||
prob = sigmoid(np.dot(inX, weights)) # 计算预测概率
|
prediction = model.predict([inX])
|
||||||
return 1 if prob > 0.5 else 0 # 返回二分类标签
|
return int(prediction[0])
|
||||||
|
|
||||||
import re
|
|
||||||
|
|
||||||
|
|
||||||
def load_data(filepath):
|
def load_data(filepath):
|
||||||
"""
|
"""
|
||||||
@@ -81,9 +62,7 @@ def load_data(filepath):
|
|||||||
data = []
|
data = []
|
||||||
with open(filepath, 'r') as file:
|
with open(filepath, 'r') as file:
|
||||||
for line in file:
|
for line in file:
|
||||||
# 使用正则表达式按空白字符分隔
|
currLine = line.strip().split('\t')
|
||||||
currLine = re.split(r"\s+", line.strip())
|
|
||||||
# 将数据转换为浮点数,如果是 "?" 则替换为 NaN
|
|
||||||
lineData = [float(x) if x != '?' else np.nan for x in currLine]
|
lineData = [float(x) if x != '?' else np.nan for x in currLine]
|
||||||
data.append(lineData)
|
data.append(lineData)
|
||||||
|
|
||||||
@@ -91,25 +70,24 @@ def load_data(filepath):
|
|||||||
X = data[:, :-1]
|
X = data[:, :-1]
|
||||||
y = data[:, -1]
|
y = data[:, -1]
|
||||||
|
|
||||||
# 使用简单的列均值填补缺失值
|
# 填充缺失值
|
||||||
col_mean = np.nanmean(X, axis=0)
|
col_mean = np.nanmean(X, axis=0)
|
||||||
inds = np.where(np.isnan(X))
|
inds = np.where(np.isnan(X))
|
||||||
X[inds] = np.take(col_mean, inds[1])
|
X[inds] = np.take(col_mean, inds[1])
|
||||||
|
|
||||||
return X, y.astype(int)
|
return X, y.astype(int)
|
||||||
|
|
||||||
|
|
||||||
if __name__ == "__main__":
|
if __name__ == "__main__":
|
||||||
# 以下代码可用于测试,可自行随意修改
|
|
||||||
# 加载训练集
|
# 加载训练集
|
||||||
X_train, y_train = load_data("./data/horse-colic.data")
|
X_train, y_train = load_data("horseColicTraining.txt")
|
||||||
|
|
||||||
# 加载测试集
|
# 训练模型
|
||||||
X_test, y_test = load_data("./data/horse-colic.test")
|
model = train_logreg(X_train, y_train, max_iter=1000)
|
||||||
|
|
||||||
# 训练对率回归模型
|
# 加载测试集,获取单个数据进行预测
|
||||||
weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000)
|
X_test, y_test = load_data("horseColicTest.txt")
|
||||||
|
test_sample = X_test[0] # 选取测试集的第一个样本
|
||||||
|
|
||||||
# 测试单个数据
|
# 预测单个样本
|
||||||
y_pred = predict_logreg(X_test[11], weights)
|
y_pred = predict_logreg(test_sample, model)
|
||||||
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[0]}")
|
||||||
|
|||||||
Reference in New Issue
Block a user