更新 function.py
Gitea Actions Experiment Test / Experiment-Actions (push) Successful in 14s

This commit is contained in:
aaa
2024-11-03 22:15:42 +08:00
parent 9666cf416b
commit 6ce5e06870
+106 -57
View File
@@ -1,81 +1,130 @@
# function.py
import numpy as np import numpy as np
import os
from sklearn.preprocessing import StandardScaler # 导入标准化库
""" # 学号和姓名
实验说明: id = "202210423122"
- 自行编写实现数据加载与预测病马死亡的对率回归算法。 name = "焦国波"
注意事项: def sigmoid(z):
- 在 function.py 中实现对率回归算法。 """计算sigmoid函数"""
- 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。 return 1.0 / (1 + np.exp(-z))
- 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: def compute_gradient(X_batch, y_batch, weights, lambda_reg):
- CPU: Intel Xeon(R) CPU E3-1231 v3 @ 3.40GHz """计算梯度,包含L2正则化"""
- OS: Ubuntu 22.04.4 LTS z = np.dot(X_batch, weights)
- Python: 3.10.12 h = sigmoid(z)
- numpy: 2.1.1 gradient = np.dot(X_batch.T, (h - y_batch)) / len(y_batch)
""" gradient += (lambda_reg / len(y_batch)) * weights # L2正则化
return gradient
def train_logreg(X, y, learning_rate=0.001, iterations=10000, batch_size=64, lambda_reg=0.1):
"""对率回归训练函数。"""
m, n = X.shape
weights = np.random.randn(n) * np.sqrt(2. / n) # He初始化
print("开始训练对率回归模型")
""" for i in range(iterations):
请在这里填写学号和姓名,不填写即成绩无效。 indices = np.random.permutation(m)
姓名请直接**填写中文** X_shuffled = X[indices]
例如: name = "张三" y_shuffled = y[indices]
"""
id = "202210423xxx"
name = "test_name"
for j in range(0, m, batch_size):
X_batch = X_shuffled[j:j + batch_size]
y_batch = y_shuffled[j:j + batch_size]
gradient = compute_gradient(X_batch, y_batch, weights, lambda_reg)
weights -= learning_rate * gradient
# 这里写自己的代码实现 if i % 1000 == 0: # 每1000次迭代打印一次信息
def train_logreg(X, y, other_params="你自己设置一些其它参数"): print(f"迭代 {i}: 当前权重形状: {weights.shape}")
"""
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,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**。
Returns:
- numpy array: 训练得到的权重。
"""
return weights
def predict_logreg(inX, weights): def predict_logreg(inX, weights):
""" """对率回归预测函数。"""
对率回归预测函数。 z = np.dot(inX, weights)
h = sigmoid(z)
return 1 if h >= 0.5 else 0
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。
"""
# 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。"""
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**。 if not os.path.exists(filepath):
raise FileNotFoundError(f"文件 {filepath} 不存在!")
Args: data, labels = [], []
- directory (str): 数据文件路径。 print(f"加载数据中: {filepath}")
with open(filepath, 'r') as f:
for line in f:
line = line.strip().split()
if len(line) >= 1 and line[-1].isdigit():
line_data = [0 if x == '?' else float(x) for x in line[:-1]]
data.append(line_data)
labels.append(int(line[-1]))
Returns: data = np.array(data)
- np.array: 特征数据。 labels = np.array(labels)
- np.array: 标签数据。
"""
# 去掉特定特征
indices_to_remove = [2] # 去掉特征3(索引2)
if data.shape[1] > 23:
indices_to_remove.extend(range(23, min(28, data.shape[1])))
data = np.delete(data, indices_to_remove, axis=1)
target_variable = np.where((labels == 2) | (labels == 3), 0, 1)
print(f"数据加载成功,数据形状: {data.shape}, 标签形状: {target_variable.shape}")
return data, target_variable
def k_fold_cross_validation(X, y, k=5, learning_rate=0.001, iterations=10000, batch_size=64, lambda_reg=0.1):
"""K折交叉验证。"""
m = X.shape[0]
fold_size = m // k
accuracies = []
for i in range(k):
val_start = i * fold_size
val_end = (i + 1) * fold_size if i != k - 1 else m
X_val = X[val_start:val_end]
y_val = y[val_start:val_end]
X_train = np.concatenate((X[:val_start], X[val_end:]))
y_train = np.concatenate((y[:val_start], y[val_end:]))
weights = train_logreg(X_train, y_train, learning_rate, iterations, batch_size, lambda_reg)
val_accuracy = np.mean([predict_logreg(X_val[j], weights) == y_val[j] for j in range(len(y_val))])
accuracies.append(val_accuracy)
print(f"Fold {i + 1}/{k} - 验证准确率: {val_accuracy:.4f}")
return np.mean(accuracies)
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") try:
X_train, y_train = load_data("./data/horse-colic.data")
except FileNotFoundError as e:
print(e)
exit()
# 数据归一化处理
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train) # 使用训练集进行标准化
# k折交叉验证
avg_accuracy = k_fold_cross_validation(X_train, y_train, k=5, learning_rate=0.01, iterations=1000)
print(f"K折交叉验证的平均准确率: {avg_accuracy:.4f}")
# 加载测试集 # 加载测试集
X_test, y_test = load_data("./data/horse-colic.test") try:
X_test, y_test = load_data("./data/horse-colic.test")
except FileNotFoundError as e:
print(e)
exit()
# 测试集也需要进行相同的归一化
X_test = scaler.transform(X_test) # 使用相同的Scaler进行测试集的标准化
# 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, learning_rate=0.01, iterations=1000)