Tensor must be from the same graph as Tensor

ValueError: Tensorflow error: “Tensor must be from the same graph as Tensor

在使用keras的backend.set_value(x, value)时出现以上报错，是由于set_value(x, value)函数定义中使用placeholder与x不属于同一图导致，修改函数定义如下即可：

def set_value(x, value):   
    value = np.asarray(value, dtype=dtype(x))
    tf_dtype = tf.as_dtype(x.dtype.name.split('_')[0])
    if hasattr(x, '_assign_placeholder'):
        assign_placeholder = x._assign_placeholder
        assign_op = x._assign_op
    else:
        assign_placeholder = tf.placeholder(tf_dtype, shape=value.shape)
        assign_op = x.assign(value)
        x._assign_placeholder = assign_placeholder
        x._assign_op = assign_op
    get_session().run(assign_op)