对TensorFlow对象执行操作时获取类型错误

下面是代码的简化版本,我在第行出错 res = input - var

import tensorflow.contrib.slim as slim
import tensorflow as tf

x = tf.placeholder(tf.float32, shape=[None, 150, 220, 3], name='x')

input = slim.conv2d(x,  num_outputs=96, kernel_size=11, stride=4, padding=padding, scope=scope, weights_initializer=tf.truncated_normal_initializer(stddev=0.01), biases_initializer=None, activation_fn=None)

var = tf.zeros_initializer()

res = input - var

变量类型

input type: <tensorflow.python.ops.init_ops.Zeros object at 0x7f1a017fdb38>
var_type: Tensor("conv1/Conv2D:0", shape=(?, 35, 53, 96), dtype=float32)

误差

TypeError: Expected float32, got <tensorflow.python.ops.init_ops.Zeros object at 0x7f1a017fdb38> of type 'Zeros' instead.

我该怎么解决呢,任何建议都会有帮助的。