tf.nn.embedding_lookup函數的用法主要是選取一個張量里面索引對應的元素。tf.nn.embedding_lookup(tensor, id):tensor就是輸入張量,id就是張量對應的索引,其他的參數不介紹。
例如:
import tensorflow as tf; import numpy as np; c = np.random.random([10,1]) b = tf.nn.embedding_lookup(c, [1, 3]) with tf.Session() as sess: sess.run(tf.initialize_all_variables()) print sess.run(b) print c
輸出:
[[ 0.77505197]
[ 0.20635818]]
[[ 0.23976515]
[ 0.77505197]
[ 0.08798201]
[ 0.20635818]
[ 0.37183035]
[ 0.24753178]
[ 0.17718483]
[ 0.38533808]
[ 0.93345168]
[ 0.02634772]]
分析:輸出為張量的第一和第三個元素。
