tensorflow之tf.nn.l2_normalize與l2_loss的計算
1.tf.nn.l2_normalize
tf.nn.l2_normalize(x, dim, epsilon=1e-12, name=None)
上式:
x為輸入的向量;
dim為l2範化的維數,dim取值為0或0或1;
epsilon的範化的最小值邊界;
按照列計算:
import tensorflow as tf input_data = tf.constant([[1.0,2,3],[4.0,5,6],[7.0,8,9]]) output = tf.nn.l2_normalize(input_data, dim = 0) with tf.Session() as sess: print sess.run(input_data) print sess.run(output)
[[1. 2. 3.]
[4. 5. 6.]
[7. 8. 9.]]
[[0.12309149 0.20739034 0.26726127]
[0.49236596 0.51847583 0.53452253]
[0.86164045 0.82956135 0.80178374]]
[[1./norm(1), 2./norm(2) , 3./norm(3) ]
[4./norm(1) , 5./norm(2) , 6./norm(3) ] =
[7./norm(1) , 8./norm(2) , 9./norm(3) ]]
[[0.12309149 0.20739034 0.26726127]
[0.49236596 0.51847583 0.53452253]
[0.86164045 0.82956135 0.80178374]]
按照行計算:
import tensorflow as tf
input_data = tf.constant([[1.0,2,3],[4.0,5,6],[7.0,8,9]])
output = tf.nn.l2_normalize(input_data, dim = 1)
with tf.Session() as sess:
print sess.run(input_data)
print sess.run(output)
[[1. 2. 3.] [4. 5. 6.] [7. 8. 9.]] [[0.26726124 0.5345225 0.8017837 ] [0.45584232 0.5698029 0.6837635 ] [0.5025707 0.5743665 0.64616233]]
[[1./norm(1), 2./norm(1) , 3./norm(1) ]
[4./norm(2) , 5./norm(2) , 6./norm(2) ] =
[7./norm(3) , 8..norm(3) , 9./norm(3) ]]
[[0.12309149 0.20739034 0.26726127]
[0.49236596 0.51847583 0.53452253]
[0.86164045 0.82956135 0.80178374]]
2.tf.nn.l2_loss
tf.nn.l2_loss(t, name=None)
解釋:這個函式的作用是利用 L2 範數來計算張量的誤差值,但是沒有開方並且只取 L2 範數的值的一半,具體如下:
output = sum(t ** 2) / 2
import tensorflow as tf
a=tf.constant([1,2,3],dtype=tf.float32)
b=tf.constant([[1,1],[2,2],[3,3]],dtype=tf.float32)
with tf.Session() as sess:
print('a:')
print(sess.run(tf.nn.l2_loss(a)))
print('b:')
print(sess.run(tf.nn.l2_loss(b)))
sess.close()
輸出結果: a: 7.0 b: 14.0
輸入引數:
- t: 一個Tensor。資料型別必須是一下之一:float32,float64,int64,int32,uint8,int16,int8,complex64,qint8,quint8,qint32。雖然一般情況下,資料維度是二維的。但是,資料維度可以取任意維度。
- name: 為這個操作取個名字。
輸出引數:
一個 Tensor ,資料型別和 t 相同,是一個標量。