how to predict with .meta and checkpoint files in tensorflow?

问题

I'm learning about MobileNet thesedays and i'm new to tensorflow. After training with ssd-mobilenet model,i got checkpoint file , .meta file , graph.pbtxt file and so on. When I try to predict with these files, i can't get the output such as box_pred, classs_scores...

Then I found predict demo code used .pb file to load graph ,and used "get_tensor_by_name" to get output, but Ｉ don't have .pb file. So, how can I predict an image with .meta and ckpt files ?

BTW, here is predict demon main code:

import numpy as np
import os
import six.moves.urllib as urllib
import sys
import tarfile
import tensorflow as tf
import zipfile
import time

from collections import defaultdict
from io import StringIO
from matplotlib import pyplot as plt
from PIL import Image

#%matplotlib inline

# This is needed since the notebook is stored in the object_detection folder.
sys.path.append("..")

from utils import label_map_util
from utils import visualization_utils as vis_util

# What model to download.
MODEL_NAME = 'ssd_mobilenet_v1_coco_11_06_2017'
MODEL_FILE = MODEL_NAME + '.tar.gz'
DOWNLOAD_BASE = 'http://download.tensorflow.org/models/object_detection/'

# Path to frozen detection graph. This is the actual model that is used for the object detection.
PATH_TO_CKPT = MODEL_NAME + '/frozen_inference_graph.pb'

# List of the strings that is used to add correct label for each box.
PATH_TO_LABELS = os.path.join('data', 'mscoco_label_map.pbtxt')

NUM_CLASSES = 90

#Load a (frozen) Tensorflow model into memory.
detection_graph = tf.Graph()
with detection_graph.as_default():
    od_graph_def = tf.GraphDef()
    with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:
        serialized_graph = fid.read()
        od_graph_def.ParseFromString(serialized_graph)
        tf.import_graph_def(od_graph_def, name='')

#load label map
label_map = label_map_util.load_labelmap(PATH_TO_LABELS)
categories = label_map_util.convert_label_map_to_categories(label_map, max_num_classes=NUM_CLASSES, use_display_name=True)
category_index = label_map_util.create_category_index(categories)

def load_image_into_numpy_array(image):
  (im_width, im_height) = image.size
  return np.array(image.getdata()).reshape(
      (im_height, im_width, 3)).astype(np.uint8)


#detection
# For the sake of simplicity we will use only 2 images:
# image1.jpg
# image2.jpg
# If you want to test the code with your images, just add path to the images to the TEST_IMAGE_PATHS.
PATH_TO_TEST_IMAGES_DIR = 'test_images'
TEST_IMAGE_PATHS = [ os.path.join(PATH_TO_TEST_IMAGES_DIR, 'image{}.jpg'.format(i)) for i in range(1, 3) ]

# Size, in inches, of the output images.
IMAGE_SIZE = (12, 8)

with detection_graph.as_default():
  with tf.Session(graph=detection_graph) as sess:
    for image_path in TEST_IMAGE_PATHS:
      image = Image.open(image_path)
      # the array based representation of the image will be used later in order to prepare the
      # result image with boxes and labels on it.
      image_np = load_image_into_numpy_array(image)
      # Expand dimensions since the model expects images to have shape: [1, None, None, 3]
      image_np_expanded = np.expand_dims(image_np, axis=0)
      image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
      # Each box represents a part of the image where a particular object was detected.
      boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
      # Each score represent how level of confidence for each of the objects.
      # Score is shown on the result image, together with the class label.
      scores = detection_graph.get_tensor_by_name('detection_scores:0')
      classes = detection_graph.get_tensor_by_name('detection_classes:0')
      num_detections = detection_graph.get_tensor_by_name('num_detections:0')
      # Actual detection.
      (boxes, scores, classes, num_detections) = sess.run(
          [boxes, scores, classes, num_detections],
          feed_dict={image_tensor: image_np_expanded})

回答1:

You should load the graph using tf.train.import_meta_graph() and then get the tensors using get_tensor_by_name(). You can try:

model_path = "model.ckpt"
detection_graph = tf.Graph()
with tf.Session(graph=detection_graph) as sess:
    # Load the graph with the trained states
    loader = tf.train.import_meta_graph(model_path+'.meta')
    loader.restore(sess, model_path)

    # Get the tensors by their variable name
    image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
    boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
    scores = detection_graph.get_tensor_by_name('detection_scores:0')
    ...
    # Make predictions
    _boxes, _scores = sess.run([boxes, scores], feed_dict={image_tensor: image_np_expanded})

回答2:

Just for those who have the problem like wu ruize and CoupDeMistral:

But I got this error: "The name 'image_tensor:0' refers to a Tensor which does not exist. The operation, 'image_tensor', does not exist in the graph."

You need to name your tensor first before using detection_graph.get_tensor_by_name.

For example, something like this:

accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32),name='accuracy')

Notice that the tensor above has been named as 'accuracy'.

After that you can enjoy the restore operation by:

detection_graph.get_tensor_by_name('accuracy:0')

Have fun now :P!

来源：https://stackoverflow.com/questions/44873204/how-to-predict-with-meta-and-checkpoint-files-in-tensorflow

标签

python

tensorflow

deep-learning

detection