张量流框架中的运行代码问题

时间:2018-10-13 00:53:49

标签: python tensorflow

tensorflow版本:1.4.0 python版本:2.7

def build_detection(self):     self.embeds,self.embeds1 = self.get_image_embedding(self.search_images,重用= True)     使用tf.variable_scope('detection'):       def _translation_match(x,z):         x = tf.expand_dims(x,0)#[批处理,in_height,in_width,in_channels]         z = tf.expand_dims(z,-1)#[filter_height,filter_width,in_channels,out_channels]         返回tf.nn.conv2d(x,z,strides = [1,1,1,1],padding ='VALID',name ='translation_match')

  output = tf.map_fn(
    lambda x: _translation_match(x[0], x[1]),
    (self.embeds, self.templates), dtype=self.embeds.dtype)  # of shape [16, 1, 17, 17, 1]
  output = tf.squeeze(output, [1, 4])  # of shape e.g. [16, 17, 17]
  output1 = tf.map_fn(
    lambda x: _translation_match(x[0], x[1]),
    (self.embeds1, self.templates1), dtype=self.embeds1.dtype)  # of shape [16, 1, 17, 17, 1]
  output1 = tf.squeeze(output1, [1, 4])

  bias = tf.get_variable('biases', [1],
                         dtype=tf.float32,
                         initializer=tf.constant_initializer(0.0, dtype=tf.float32),
                         trainable=False)
  response = self.model_config['adjust_response_config']['scale'] * output + bias
  response1 = self.model_config['adjust_response_config']['scale'] * output1 + bias
  self.response = response, self.response1 = response1

restore_fn = model.build_graph_from_config(model_config,track_config,检查点)   文件“ /home/xieyangyang/Downloads/SiamFC-TensorFlow-master/scripts/inference/inference_wrapper.py”,第51行,位于build_graph_from_config中     self.build_model(model_config,track_config)   在build_model中的文件“ /home/xieyangyang/Downloads/SiamFC-TensorFlow-master/scripts/inference/inference_wrapper.py”,第82行     self.build_detection()   在build_detection中的文件“ /home/xieyangyang/Downloads/SiamFC-TensorFlow-master/scripts/inference/inference_wrapper.py”中,第229行     self.response =响应,self.response1 = response1   在 iter 中的第505行中的文件“ /home/xieyangyang/anaconda3/envs/tensorflow1/lib/python2.7/site-packages/tensorflow/python/framework/ops.py”     引发TypeError(“'Tensor'对象不可迭代。”) TypeError:“张量”对象不可迭代。

0 个答案:

没有答案