tf 实现 KNN

mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)     #下载并加载mnist数据

train_X, train_Y = mnist.train.next_batch(5000) # 5000 for training (nn candidates)
test_X, test_Y = mnist.test.next_batch(100)   # 200 for testing


tra_X = tf.placeholder("float", [None, 784])
te_X = tf.placeholder("float", [784])

# Nearest Neighbor calculation using L1 Distance
# Calculate L1 Distance
distance = tf.reduce_sum(tf.abs(tf.add(tra_X, tf.negative(te_X))), reduction_indices=1)
# Prediction: Get min distance index (Nearest neighbor)
pred = tf.arg_min(distance, 0)

accuracy = 0.

# Initializing the variables
init = tf.initialize_all_variables()

# Launch the graph
with tf.Session() as sess:
  sess.run(init)

    # loop over test data
  for i in range(len(test_X)):
    nn_index = sess.run(pred, feed_dict={tra_X: train_X, te_X: test_X[i, :]}) # Get nearest neighbor
    print("Test", i, "Prediction:", np.argmax(train_Y[nn_index]), "True Class:", np.argmax(test_Y[i])) # Get nearest neighbor class label and compare it to its true label
    if np.argmax(train_Y[nn_index]) == np.argmax(test_Y[i]): # Calculate accuracy
      accuracy += 1./len(test_X)
  print("Done!")
  print("Accuracy:", accuracy)