python - Tensor multiplication in Tensorflow -

i trying carry out tensor multiplication in numpy/tensorflow.

i have 3 tensors- a (m x h), b (h x n x s), c (s x t).

i believe a x b x c should produce tensor d (m x n x t).

here's code (using both numpy , tensorflow).

m = 5 n = 2 t = 3 h = 2 s = 3 a_np = np.random.randn(m, h) c_np = np.random.randn(s, t) b_np = np.random.randn(h, n, s)  a_tf = tf.variable(a_np) c_tf = tf.variable(c_np) b_tf = tf.variable(b_np)  # tensorflow tf.session() sess:     sess.run(tf.global_variables_initializer())     print sess.run(a_tf)     p = tf.matmul(a_tf, b_tf)     sess.run(p) 

this returns following error:

valueerror: shape must rank 2 rank 3 'matmul_2' (op: 'matmul') input shapes: [5,2], [2,2,3]. 

if try multiplication numpy matrices, following errors:

np.multiply(a_np, b_np)  valueerror: operands not broadcast shapes (5,2) (2,2,3) 

however, can use np.tensordot follows:

np.tensordot(np.tensordot(a_np, b_np, axes=1), c_np, axes=1) 

is there equivalent operation in tensorflow?

answer

in numpy, follows:

abc_np = np.tensordot(np.tensordot(a_np, b_np, axes=1), c_np, axes=1) 

in tensorflow, follows:

ab_tf = tf.tensordot(a_tf, b_tf,axes = [[1], [0]]) ab_tf_c_tf = tf.tensordot(ab_tf, c_tf, axes=[[2], [0]])  tf.session() sess:     sess.run(tf.global_variables_initializer())     abc_tf = sess.run(ab_tf_c_tf) 

np.allclose(abc_np, abc_tf) return true.

try

tf.tensordot(a_tf, b_tf,axes = [[1], [0]]) 

for example:

x=tf.tensordot(a_tf, b_tf,axes = [[1], [0]]) x.get_shape() tensorshape([dimension(5), dimension(2), dimension(3)]) 

here tensordot documentation, , here relevant github reciprocity.