Classification of MNIST with scattering

Here we demonstrate a simple application of scattering on the MNIST dataset. We use 10000 images to train a linear classifier. Features are normalized by batch normalization.

Preliminaries

Since we’re using TensorFlow and Keras to train the model, import the relevant modules.

import tensorflow as tf
from tensorflow.keras.models import Model
from tensorflow.keras.layers import Input, Flatten, Dense

Finally, we import the Scattering2D class from the kymatio.keras package.

from kymatio.keras import Scattering2D

Training and testing the model

First, we load in the data and normalize it.

(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()

x_train, x_test = x_train / 255., x_test / 255.

We then create a Keras model using the scattering transform followed by a dense layer and a softmax activation.

inputs = Input(shape=(28, 28))
x = Scattering2D(J=3, L=8)(inputs)
x = Flatten()(x)
x_out = Dense(10, activation='softmax')(x)
model = Model(inputs, x_out)

Display the created model.

model.summary()

Once the model is created, we couple it with an Adam optimizer and a cross-entropy loss function.

model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

We then train the model using model.fit on a subset of the MNIST data.

model.fit(x_train[:10000], y_train[:10000], epochs=15,
          batch_size=64, validation_split=0.2)

Finally, we evaluate the model on the held-out test data.

model.evaluate(x_test, y_test)