nu = np.array([[[1, 1, 1, 1],
[2, 2, 2, 2],
[3, 3, 3, 3]],
[[1, 1, 1, 1],
[2, 2, 2, 2],
[3, 3, 3, 3]]])
mu = nu.reshape([1,2,3,4])
div = tf.reduce_min(mu, axis=3, keepdims=True)
with tf.Session() as sess:
d = sess.run(div)
运行结果
版权声明：本文为weixin_41855385原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。