tf.nn.embedding_lookup(params, ids, partition_strategy='mod', name=None, validate_indices=True, max_norm=None)

首先通过一下一个简单的例子来了解一下tf.nn.embedding_lookup()的用法

a = tf.constant([[1,2,3],[4,5,6],[7,8,9],[10,11,12]])

>>>a.eval()

Out[51]:

array([[ 1, 2, 3],

[ 4, 5, 6],

[ 7, 8, 9],

[10, 11, 12]], dtype=int32)

 
 

tf.nn.embedding_lookup(a,[0,1]).eval()

>>>Out[52]:

array([[1, 2, 3],

[4, 5, 6]], dtype=int32)

 
 

tf.nn.embedding_lookup(a,[[0,1],[1,2]]).eval()

>>>Out[53]:

array([[[1, 2, 3],

[4, 5, 6]],

[[4, 5, 6],

[7, 8, 9]]], dtype=int32)