用python代码实现使用KNN算法识别mnist.test_image中的每个图片对应的数值并于mnist.test_labels进行比较看是否正确。已有mnist.train_image和mnist.test_labels

首先，需要导入必要的库和数据集：

import numpy as np
from sklearn.neighbors import KNeighborsClassifier
from tensorflow.examples.tutorials.mnist import input_data

mnist = input_data.read_data_sets("MNIST_data/", one_hot=False)

然后，将训练集和测试集的图像数据和标签提取出来：

train_images = mnist.train.images
train_labels = mnist.train.labels
test_images = mnist.test.images
test_labels = mnist.test.labels

接下来，我们可以使用KNN算法进行训练和预测：

knn = KNeighborsClassifier(n_neighbors=3)
knn.fit(train_images, train_labels)

predicted_labels = knn.predict(test_images)

最后，我们可以比较预测的标签和测试集中的真实标签，计算模型的准确率：

accuracy = np.mean(predicted_labels == test_labels)
print("Accuracy:", accuracy)

完整的代码如下：

import numpy as np
from sklearn.neighbors import KNeighborsClassifier
from tensorflow.examples.tutorials.mnist import input_data

mnist = input_data.read_data_sets("MNIST_data/", one_hot=False)

train_images = mnist.train.images
train_labels = mnist.train.labels
test_images = mnist.test.images
test_labels = mnist.test.labels

knn = KNeighborsClassifier(n_neighbors=3)
knn.fit(train_images, train_labels)

predicted_labels = knn.predict(test_images)

accuracy = np.mean(predicted_labels == test_labels)
print("Accuracy:", accuracy)