python - initialize matrices in tensorflow -
i have 6 matrices, model learn, defined them follow:
self.r= tf.get_variable('r_',dtype=tf.float32, shape=[6,300 ,300], initializer=tf.random_uniform_initializer(maxval=0.1, minval=-0.1))
what need change initialization. want initialize each 1 of them identity matrix. can me that?
if want create 6x300x300 matrix each 300x300 array identity matrix can simply:
import numpy np; dimension = 300 singleidentitymatrix = np.identity(dimension, dtype= np.float32) stackedmatrix = np.dstack( [singleidentitymatrix] * 6)
and pass matrix
self.r = tf.variable(initial_value = stackedmatrix)
Comments
Post a Comment