【TensorFlow】使用TensorFlow执行K-Means

import numpy as np
import tensorflow as tf
from tensorflow.contrib.factorization import KMeans

加载数据

from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("/tmp/data/", one_hot=True)
full_data_x = mnist.train.images

Extracting /tmp/data/train-images-idx3-ubyte.gz
Extracting /tmp/data/train-labels-idx1-ubyte.gz
Extracting /tmp/data/t10k-images-idx3-ubyte.gz
Extracting /tmp/data/t10k-labels-idx1-ubyte.gz

参数

num_features = 784 # 图片尺寸为28*28=784
num_classes = 10 # 0~9共10个数字
k = 25
num_steps = 50 # 训练执行的次数
batch_size = 1024

构建模型

X = tf.placeholder(tf.float32,shape=[None,num_features])
Y = tf.placeholder(tf.float32,shape=[None,num_classes])
kmeans = KMeans(inputs=X,num_clusters=k,distance_metric='cosine',use_mini_batch=True)

(all_scores, cluster_idx, scores, cluster_centers_initialized, 
 init_op,training_op) = kmeans.training_graph()
cluster_idx = cluster_idx[0]
avg_distance = tf.reduce_mean(scores)

init_vars = tf.global_variables_initializer()

训练

sess = tf.Session()

sess.run(init_vars, feed_dict={X: full_data_x})
sess.run(init_op, feed_dict={X: full_data_x})

for i in range(1,num_steps+1):
    _,d,idx = sess.run([training_op,avg_distance,cluster_idx],feed_dict={X:full_data_x})
    if(i%10==0 or i==1):print("Step %i,Avg Distance:%f"%(i,d))

Step 1,Avg Distance:0.341471
Step 10,Avg Distance:0.221609
Step 20,Avg Distance:0.220328
Step 30,Avg Distance:0.219776
Step 40,Avg Distance:0.219419
Step 50,Avg Distance:0.219154

评估

# k个簇中，各个类型的个数
# 例如counts[i][j]是第2个簇中第j个类别样本的数量
counts = np.zeros(shape=(k,num_classes))
for i in range(len(idx)):
    # idx是所有样本所属簇的id
    counts[idx[i]] += mnist.train.labels[i]
# labels_map是一个len为25的list
# labels_map[i]表示第i簇中样本应该属于的类别
labels_map = [np.argmax(c) for c in counts]
labels_map = tf.convert_to_tensor(labels_map)

# 给定一个cluster_idx，返回这个cluster在样本中对应的labels
cluster_label = tf.nn.embedding_lookup(labels_map, cluster_idx)
correct_prediction = tf.equal(cluster_label, tf.cast(tf.argmax(Y, 1), tf.int32))
accuracy_op = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

test_x, test_y = mnist.test.images, mnist.test.labels
print("Test Accuracy:", sess.run(accuracy_op, feed_dict={X: test_x, Y: test_y}))

Test Accuracy: 0.7127