TF Keras how to get expected input shape when loading a model?

Question

Is it possbible to get the expected input shape from a 'model.h5' file? I have two models for the same dataset but with different options and shapes. The first one expects a dim of (None, 64, 48, 1) and the seconds model need input shape (None, 128, 96, 3). (Note: The width or the height are not fixed and can change when I train again). The channels problem was easy to "fix" (or bypass rather) by just using try: and except because there are only two options (1 for grayscale image and 3 for rgb image):

        channels = self.df["channels"][0]
        file = ""
        try:
            images, src_images, data = self.get_images()
            images = self.preprocess_data(images, channels)
            predictions, file = self.load_model(images, file)
            self.predict_data(src_images, predictions, data)
        except:
            if channels == 1:
                print("Except channels =", channels)
                channels = 3
                images, src_images, data = self.get_images()
                images = self.preprocess_data(images, channels)
                predictions = self.load_model(images, file)
                self.predict_data(src_images, predictions, data)
            else:
                channels = 1
                print("Except channels =", channels)
                images, src_images, data = self.get_images()
                images = self.preprocess_data(images, channels)
                predictions = self.load_model(images, file)
                self.predict_data(src_images, predictions, data)

This workaround however cannot be used for the width and height of an image because there basically unlimited amount of options. Besides that it is rather slow because I read all the data twice and preprocess it twice for no reason.

Is there a way to load the model.h5 file and print the expected input shape in a form like this?:

[None, 128, 96, 3]

Daniel · Accepted Answer · 2021-01-23 18:47:40Z

19

I finally found the answer myself.

config = model.get_config() # Returns pretty much every information about your model
print(config["layers"][0]["config"]["batch_input_shape"]) # returns a tuple of width, height and channels

This will output the following:

(None, 128, 96, 3)

answered Jan 23, 2021 at 18:47

Daniel

1,0252 gold badges16 silver badges34 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

AnthonyT Over a year ago

Thanks. I dont understand why .getconfig() isn't documented here...: tensorflow.org/api_docs/python/tf/keras/Model

DarkHark Over a year ago

You can also use .summary() to see the shape of the input layers, among many other details.

Arka Mukherjee Over a year ago

Well, there is a caveat. This won't help you if your model takes in multiple inputs, as the indices can often re-direct you to embeddings or other layers.

Alex Westbrook · Accepted Answer · 2023-02-10 14:08:23Z

0

I found the answer from here to be more concise:

    model.layers[0].input_shape[0]

The way I understand it, it should be easier to deal with multiple inputs that way too.

edited Feb 10, 2023 at 14:08

answered Feb 10, 2023 at 14:05

Alex Westbrook

12 bronze badges

Collectives™ on Stack Overflow

TF Keras how to get expected input shape when loading a model?

2 Answers 2

3 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related