python-kerasHow can I use Python and Keras to build a model for the MNIST dataset?

1. First, import the necessary libraries for the task:

import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D
from keras import backend as K

2. Then, load the MNIST dataset from Keras:

(x_train, y_train), (x_test, y_test) = mnist.load_data()

3. Next, reshape the data to fit the model:

x_train = x_train.reshape(x_train.shape[0], 28, 28, 1)
x_test = x_test.reshape(x_test.shape[0], 28, 28, 1)
input_shape = (28, 28, 1)

4. Then, normalize the data:

x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255

5. After that, one-hot encode the labels:

y_train = keras.utils.to_categorical(y_train, 10)
y_test = keras.utils.to_categorical(y_test, 10)

6. Finally, build and compile the model:

model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=input_shape))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dense(10, activation='softmax'))
model.compile(loss=keras.losses.categorical_crossentropy, optimizer=keras.optimizers.Adadelta(), metrics=['accuracy'])

7. Finally, fit the model:

model.fit(x_train, y_train, batch_size=128, epochs=10, verbose=1, validation_data=(x_test, y_test))

Output example

Train on 60000 samples, validate on 10000 samples
Epoch 1/10
60000/60000 [==============================] - 15s 252us/step - loss: 0.2796 - accuracy: 0.9187 - val_loss: 0.1162 - val_accuracy: 0.9648
Epoch 2/10
60000/60000 [==============================] - 5s 81us/step - loss: 0.0911 - accuracy: 0.9737 - val_loss: 0.0791 - val_accuracy: 0.9761
...

This is an example of how to use Python and Keras to build a model for the MNIST dataset. The code is composed of the following parts:

1. Import the necessary libraries for the task: keras, mnist, Sequential, Dense, Dropout, Flatten, Conv2D, MaxPooling2D and K.
2. Load the MNIST dataset from Keras.
3. Reshape the data to fit the model.
4. Normalize the data.
5. One-hot encode the labels.
6. Build and compile the model.
7. Fit the model.

Helpful links

• Keras Documentation
• MNIST Dataset

Edit this code on GitHub