tensorflow Shapes and Shaping

tf.shape(input, name=None)

mb = tf.constant([[1,2,3],[4,5,6]])
shape = tf.shape(mb)
sess.run(shape)

输出是：array([2, 3], dtype=int32)

tf.size(input, name=None)

Returns the size of a tensor.

mb = tf.constant([[1,2,3],[4,5,6]])
shape = tf.shape(mb)
shape = tf.size(mb)

输出是：6

tf.rank(input, name=None)
Returns the rank of a tensor.

注解：其中这个rank（阶），可以理解为数组的维度（几维数据）

mb = tf.constant([[1,2,3],[4,5,6]])
sess.run(tf.rank(mb)

tf.reshape(tensor, shape, name=None

Reshapes a tensor

mb = tf.constant([[1,2,3],[4,5,6]])
sess.run(tf.reshape(mb, [3,2]))

输出是：

array([[1, 2],
       [3, 4],
       [5, 6]], dtype=int32)

tf.squeeze(input, squeeze_dims=None, name=None)

Removes dimensions of size 1 from the shape of a tensor.

arr = tf.Variable(tf.truncated_normal([3,4,1,6,1], stddev=0.1))
arr2 = tf.squeeze(arr)

输出是：
array([3, 4, 6], dtype=int32

tf.expand_dims(input, dim, name=None)
Inserts a dimension of 1 into a tensor's shape.

注释：在tensor中增加维度，其中dim表示增加的位置

ma = tf.constant([1.1, 1.2])
tf.expand_dims(ma, 1)
sess.run(mc)

输出是：
array([[ 1.10000002],
       [ 1.20000005]], dtype=float32)