keras.utils importError in Colab cannot import name "to_categorical"

Question

I'm using Google's Colab to run the Deep Learning codes from the Book " Deep Learning with python" by François Chollet. The 1st exercise is to use the mnist dataset. I get the following error:

ImportError: cannot import name 'to_categorical' from 'keras.utils' (/usr/local/lib/python3.7/dist-packages/keras/utils/__init__.py)

I copied the code to make sure there are no typos.

Colab suggests "... manually install dependencies using either !pip or !apt.", in case a package is missing, which doesn't seem to be the case.

I nevertheless tried:

!pip install keras

which is either incorrect or irrelevant; the error persists.

What am I doing wrong? How can I correct it?

The code causing the error is:

from keras.utils import to_categorical

And this is the complete program directly from the book's source code.

    from keras.datasets import mnist
    from keras.utils import to_categorica
    from jeras import models
    from keras import layers

    (train_images, train_labels), (test_images, test-labels) = mnist.load_data()

    network = models.Sequential()
    network.add(layers.Dense(512, activation="relu", input_shape=(28*28,)))
    network.add(layers.Dense(10, activation="softmax"))

    network.compile(optimizer="rmsprop",
                loss="categorical_crossentropy",
                metrics=["accuracy"])

    train_images = train_images.reshape((60000, 28*28))
    train_images = train_images.astype("float32") / 255

    test_images = test_images.reshape((10000, 28*28))
    test_images = test_images.astype("float32") / 255

    train_labels = to_categorical(train_labels)
    test_labels = to_categorical(test_labels)

    network.fit(train_images, train_labels, epochs=5, batch_size=128)

    test_loss, test_acc = network.evaluate(test_images, test-test_labels)
    print("test_acc:", test_acc)

According to the book, this is "to classify grayscale images of handwritten images" and is supposedly the "Hello World" of deep learning.

Answer 1

Change the line

from keras.utils import to_categorical

to

from tensorflow.keras.utils import to_categorical

and it'll work.