tensorflow1.0でmulの代替関数

Question

tensorflow1.0では mul は使えないのですが
代わりとなる関数は matmul で完全に問題ないでしょうか？

何かtensorflow0.x のmul との違いは無いでしょうか？

Answer 1

要素毎の乗算tf.mulの代替は行列の乗算tf.matmulではなくtf.multiplyではないでしょうか？

tf.multiply(r1.0)

Returns x * y element-wise.

NOTE: tf.multiply supports broadcasting. More about broadcasting here
Args:
x: A Tensor. Must be one of the following types: half, float32, float64, uint8, int8, uint16, int16, int32, int64, complex64, complex128.
y: A Tensor. Must have the same type as x.
name: A name for the operation (optional).
Returns:
A Tensor. Has the same type as x.

tf.mul(v0.12)

Returns x * y element-wise.

NOTE: Mul supports broadcasting. More about broadcasting here
Args:
x: A Tensor. Must be one of the following types: half, float32, float64, uint8, int8, uint16, int16, int32, int64, complex64, complex128.
y: A Tensor. Must have the same type as x.
name: A name for the operation (optional).
Returns:
A Tensor. Has the same type as x.

例

import tensorflow as tf
print(tf.__version__)
a = tf.constant([1, 2, 3, 4], shape=[2, 2])
b = tf.constant([5, 6, 7, 8], shape=[2, 2])
#mu = tf.mul(a, b)      # AttributeError: module 'tensorflow' has no attribute 'mul'
mt = tf.multiply(a, b) # 要素毎に乗算
mm = tf.matmul(a, b)   # 行列の乗算
with tf.Session() as sess:
    print(a.eval())
    print(b.eval())
    print(mt.eval())
    print(mm.eval())

結果

1.0.0
[[1 2]
 [3 4]]
[[5 6]
 [7 8]]
[[ 5 12]
 [21 32]]
[[19 22]
 [43 50]]