model.fit(train_images, train_labels, validation_data=(test_images, test_labels), epochs=5)


Train on 60000 samples, validate on 10000 samples
Epoch 1/5
60000/60000 [==============================] - 40s 660us/step - loss: 0.2486 - acc: 0.9273 - val_loss: 0.1027 - val_acc: 0.9660
Epoch 2/5
60000/60000 [==============================] - 38s 638us/step - loss: 0.0853 - acc: 0.9739 - val_loss: 0.0666 - val_acc: 0.9791
Epoch 3/5
60000/60000 [==============================] - 38s 632us/step - loss: 0.0614 - acc: 0.9813 - val_loss: 0.0513 - val_acc: 0.9840
Epoch 4/5
60000/60000 [==============================] - 38s 631us/step - loss: 0.0496 - acc: 0.9842 - val_loss: 0.0429 - val_acc: 0.9870
Epoch 5/5
60000/60000 [==============================] - 38s 638us/step - loss: 0.0423 - acc: 0.9867 - val_loss: 0.0373 - val_acc: 0.9882
<keras.callbacks.History at 0x7f7a5d883668>