线性分类python代码

以下是一个简单的线性分类器的Python代码示例：

import numpy as np

class LinearClassifier:
    def __init__(self):
        self.w = None

    def train(self, X, y, learning_rate=1e-3, num_iters=1000, batch_size=200, verbose=False):
        num_train, dim = X.shape
        num_classes = np.max(y) + 1
        if self.w is None:
            self.w = 0.001 * np.random.randn(dim, num_classes)

        for i in range(num_iters):
            batch_indices = np.random.choice(num_train, batch_size)
            X_batch = X[batch_indices]
            y_batch = y[batch_indices]

            loss, grad = self.loss(X_batch, y_batch)
            self.w -= learning_rate * grad

            if verbose and i % 100 == 0:
                print('iteration %d / %d: loss %f' % (i, num_iters, loss))

    def predict(self, X):
        y_pred = np.argmax(X.dot(self.w), axis=1)
        return y_pred

    def loss(self, X_batch, y_batch, reg=0.1):
        scores = X_batch.dot(self.w)
        num_train = X_batch.shape[0]

        correct_class_scores = scores[np.arange(num_train), y_batch]
        margins = np.maximum(0, scores - correct_class_scores[:, np.newaxis] + 1)
        margins[np.arange(num_train), y_batch] = 0
        loss = np.sum(margins) / num_train + 0.5 * reg * np.sum(self.w * self.w)

        num_pos = np.sum(margins > 0, axis=1)
        grad = np.zeros_like(self.w)
        grad[margins > 0] = 1
        grad[np.arange(num_train), y_batch] -= num_pos
        grad /= num_train
        grad += reg * self.w

        return loss, grad

这个代码实现了一个简单的线性分类器，并使用随机梯度下降法进行训练。在训练过程中，使用了随机采样的小批量数据来计算损失函数和梯度，并更新模型参数。最终，该模型可以用来对新的数据进行分类预测。