How can I use the Tensorflow .pb file?

Question

I have a Tensorflow file AlexNet.pb. I am trying to load it then classify an image that I have. I can't find a way to load it then classify an image.

No-one seems to have a simple example of loading and running the .pb file.

Possible duplicate of [Tensorflow: how to save/restore a model?](https://stackoverflow.com/questions/33759623/tensorflow-how-to-save-restore-a-model) — E_net4, Apr 23 '19 at 13:11

score 5 · Answer 1 · answered Jul 12 '18 at 08:24

It depends on how the protobuf file has been created.

If the .pb file is the result of:

    # Create a builder to export the model
    builder = tf.saved_model.builder.SavedModelBuilder("export")
    # Tag the model in order to be capable of restoring it specifying the tag set
    builder.add_meta_graph_and_variables(sess, ["tag"])
    builder.save()

You have to know how that model has been tagged and use the tf.saved_model.loader.load method to load the saved graph in the current, empty, graph.

If the model instead has been frozen you have to load the binary file in memory manually:

with tf.gfile.GFile(frozen_graph_filename, "rb") as f:
    graph_def = tf.GraphDef()
    graph_def.ParseFromString(f.read())

graph = tf.get_default_graph()
tf.import_graph_def(graph_def, name="prefix")

In both cases, you have to know the name of the input tensor and the name of the node you want to execute:

If, for example, your input tensor is a placeholder named batch_ and the node you want to execute is the node named dense/BiasAdd:0 you have to

    batch = graph.get_tensor_by_name('batch:0')
    prediction = restored_graph.get_tensor_by_name('dense/BiasAdd:0')

    values = sess.run(prediction, feed_dict={
        batch: your_input_batch,
    })

Thanks nessuno. Do you know what `.pb` stands for in TensorFlow? How is it different from using `.h5` (HDF5, a binary format) to store TensorFlow models? — Josh, Jul 31 '20 at 03:49

score -1 · Answer 2 · edited Apr 23 '19 at 12:13

-1

You can use opencv to load .pb models, eg.

net = cv2.dnn.readNet("model.pb")

Make sure you are using specific version of opencv - OpenCV 3.4.2 or OpenCV 4

edited Apr 23 '19 at 12:13

piet.t

11,718
21
43
52

answered Apr 23 '19 at 11:20

Ankit Dwivedi

27
3

1

Using OpenCV to read .pb files is not the way. OpenCV is an image library, though it started giving support for the ML/DL models. The question is better answered in with respect to Tensorflow rather than OpenCV. – Abhishek Jain Feb 03 '20 at 13:05

How can I use the Tensorflow .pb file?

2 Answers2