更新 function.py

function.py

@@ -3,31 +3,30 @@ import numpy as np
 id = "202210423201"
 name = "陈瑞琪祥"
 
-def train_svm(X, y, other_params=""):
+def train_svm(X, y, learning_rate=0.01, iterations=1000):
     """
     支持向量机训练函数。
     
     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。
+    - learning_rate (float): 学习率。
+    - iterations (int): 迭代次数。
     
     Returns:
     - numpy array: 训练得到的权重。
     """
     # 初始化权重
-    weights = np.zeros(X.shape[1])
-    # 设置学习率和迭代次数
-    learning_rate = 0.01
-    iterations = 1000
-    
+    weights = np.zeros(X.shape[1] + 1)  # 加1是为了偏置项
     # 简单的梯度下降优化
     for _ in range(iterations):
         for i in range(len(X)):
-            if y[i] * (np.dot(X[i], weights)) < 1:
-                weights += learning_rate * (y[i] * X[i])
+            if y[i] * (np.dot(X[i], weights[:-1]) + weights[-1]) < 1:
+                weights[:-1] += learning_rate * (y[i] * X[i])  # 更新权重
+                weights[-1] += learning_rate * y[i]  # 更新偏置项
             else:
-                weights -= learning_rate * (y[i] * X[i])
+                weights[:-1] -= learning_rate * (y[i] * X[i])  # 更新权重
+                weights[-1] -= learning_rate * y[i]  # 更新偏置项
     
     return weights
 
@@ -43,7 +42,7 @@ def predict_svm(inX, weights):
     - int: 预测的分类标签。
     """
     # 预测标签
-    prediction = np.sign(np.dot(inX, weights))
+    prediction = np.sign(np.dot(inX, weights[:-1]) + weights[-1])  # 加上偏置项
     return 1 if prediction > 0 else -1
 
 def load_data(filepath):