更新 function.py

function.py

@@ -9,21 +9,14 @@ def sigmoid(z):
 def train_logreg(X, y, numIter=1000, learning_rate=0.01, lambda_reg=0.01):
     m, n = X.shape
     weights = np.zeros(n)
-    val_X, val_y = X[:int(0.2*m)], y[:int(0.2*m)]  # Split 20% of data for validation
-    for i in range(numIter):
+    for j in range(numIter):
         z = np.dot(X, weights)
         h = sigmoid(z)
-        gradient = np.dot(X.T, (h - y)) / m
+        error = y - h
+        gradient = np.dot(X.T, error) / m
         reg_term = lambda_reg * weights
         gradient += reg_term / m
         weights -= learning_rate * gradient
-        # Early stopping
-        if i % 100 == 0:  # Check every 100 iterations
-            val_z = np.dot(val_X, weights)
-            val_h = sigmoid(val_z)
-            val_error = val_h - val_y
-            if np.mean(np.abs(val_error)) < 0.01:  # If the validation error is small enough
-                break
     return weights
 
 def predict_logreg(inX, weights):
@@ -54,8 +47,8 @@ def load_data(filepath):
     data_number = len(y)
     y_train = np.array([1 if y[j] == '2' or y[j] == '3' else 0 for j in range(data_number)])
     X = np.array(data_list)
-    X_train = np.insert(X, 0, 1, axis=1)  # 添加偏置项
-    return X_train, y_train
+    X = np.insert(X, 0, 1, axis=1)  # 添加偏置项，应该在第一列添加
+    return X, y_train
 
 def standardize(X):
     mean = X.mean(axis=0)