TensorFlow slim inception resnet v2 推理

Posted 2023-03-15

【中文标题】TensorFlow slim inception resnet v2 推理

【英文标题】：Tensorflow slim inception resnet v2 inference

【发布时间】：2018-11-08 01:05:46

【问题描述】：

https://github.com/tensorflow/models/blob/master/research/slim/nets/inception_resnet_v2.py

您如何正确地对此模型进行推理？我有兴趣设置它，以便用户可以对他输入到命令行中的单个图像进行推理。为了加快速度，模型必须加载一次，并且输入图像必须是可热交换的，因为用户将它们输入到命令行中。

如果您使用与此模型的评估代码类似的结构，则可以进行非热插拔推理：https://github.com/tensorflow/models/blob/master/research/slim/eval_image_classifier.py

您可以稍微修改上述文件以打印您的 logits 并进行推理。但是，此解决方案每次都会重建图形，这确实很慢。

我尝试构建图表并将 feed_dict 传递给 fifo_queue_Dequeue:0 张量，它表示批处理输入。但是，会话将挂起并且永远不会计算。我相信这是因为图表被“冻结”了——张量不能接受新的输入。但现在我对如何获得我想要的行为感到困惑。

【参考方案1】：

推理步骤如下：

创建 Inception-resnet-v2 图

import sys
# import from tensorflow models
sys.path.append('/home/vijay/workspace/learning/tensorflow/')

#Load the definitions of Inception-Resnet-v2 architecture
import tensorflow.contrib.slim as slim
from models.research.slim.nets.inception_resnet_v2 import inception_resnet_v2, inception_resnet_v2_arg_scope


#The pretrained model accepts size of 299x299 images
HEIGHT = 299
WIDTH = 299
CHANNELS = 3

# Create Graph

graph = tf.Graph()
with graph.as_default():

   # Create a placeholder to pass the input image
   img_tensor = tf.placeholder(tf.float32, shape=(None, HEIGHT, WIDTH, CHANNELS))

   # Scale the image inputs to +1, -1 from 0 to 255
   img_scaled = tf.scalar_mul((1.0/255), img_tensor)
   img_scaled = tf.subtract(img_scaled, 0.5)
   img_scaled = tf.multiply(img_scaled, 2.0)

   # load Graph definitions
   with slim.arg_scope(inception_resnet_v2_arg_scope()):
      logits, end_points = inception_resnet_v2(img_scaled, is_training=False)

   # predict the class
   predictions = end_points['Predictions']

加载测试图像（来自here的示例）：

#Loading a test image 
img = cv2.imread('/home/vijay/datasets/image/misc/Bernese-Mountain- Dog.jpg')
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
img = cv2.resize(img, (WIDTH, HEIGHT))

# make the input size [BATCH, WIDTH, HEIGHT, CHANNELS] for the network
img = np.expand_dims(img, axis=0)

加载权重并运行图表

#for labels of imagenet 
sys.path.append('/home/vijay/workspace/learning/tensorflow/models/research/slim')
from datasets import imagenet

# Inception resnet v2 model 
checkpoint_file='/home/vijay/datasets/pre_trained_models/inception_resnet_v2_2016_08_30.ckpt'

with tf.Session(graph=train_graph) as sess:

   saver = tf.train.Saver()
   saver.restore(sess, checkpoint_file)

   pred_prob= sess.run(predictions, feed_dict=img_tensor:img)

   # Getting the top 5 classes of the imagenet database
   probabilities = pred_prob[0, 0:]
   sorted_inds = [i[0] for i in sorted(enumerate(-probabilities), key=lambda x:x[1])]

   names = imagenet.create_readable_names_for_imagenet_labels()
   for i in range(5):
       index = sorted_inds[i]
       print('Probability %0.2f%% => [%s]' % (probabilities[index], names[index]))

输出

Probability 0.84% => [Bernese mountain dog]
Probability 0.04% => [Appenzeller]
Probability 0.03% => [EntleBucher]
Probability 0.01% => [Greater Swiss Mountain dog]
Probability 0.00% => [Border collie]

【讨论】：

我尝试了此代码的类似版本：github.com/richardrl/acme-slim/blob/master/… 我收到此错误：tensorflow.python.framework.errors_impl.InvalidArgumentError: You must feed a value for placeholder tensor 'Placeholder' with dtype uint8和 shape [500,500,3] [[Node: Placeholder = Placeholder[dtype=DT_UINT8, shape=[500,500,3], _device="/job:localhost/replica:0/task:0/device:GPU:0"] ()]]

dtype 和 shape 是正确的，但错误表明它们不正确。您对如何调试有任何想法吗？

链接失效。以上代码为批处理演示，需要创建img和[batch, 299, 299, 3]

抱歉，您能再检查一下吗？

您输入的内容是什么？应该是从图像文件解码的uint8中大小为 [500, 500,3] 的 numpy 数组。