Model: "sequential_2"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
dense_6 (Dense) (None, 256) 200960
_________________________________________________________________
activation_6 (Activation) (None, 256) 0
_________________________________________________________________
dense_7 (Dense) (None, 256) 65792
_________________________________________________________________
activation_7 (Activation) (None, 256) 0
_________________________________________________________________
dense_8 (Dense) (None, 10) 2570
_________________________________________________________________
activation_8 (Activation) (None, 10) 0
=================================================================
Total params: 269,322
Trainable params: 269,322
Non-trainable params: 0
_________________________________________________________________
Train on 60000 samples
Epoch 1/20
60000/60000 [==============================] - 3s 57us/sample - loss: 1.1975 - accuracy: 0.7223
Epoch 2/20
60000/60000 [==============================] - 3s 43us/sample - loss: 0.4732 - accuracy: 0.8772
Epoch 3/20
60000/60000 [==============================] - 3s 44us/sample - loss: 0.3731 - accuracy: 0.8975
Epoch 4/20
60000/60000 [==============================] - 2s 37us/sample - loss: 0.3302 - accuracy: 0.9069
Epoch 5/20
60000/60000 [==============================] - 3s 44us/sample - loss: 0.3032 - accuracy: 0.9145
Epoch 6/20
60000/60000 [==============================] - 2s 38us/sample - loss: 0.2829 - accuracy: 0.9197
Epoch 7/20
60000/60000 [==============================] - 2s 38us/sample - loss: 0.2666 - accuracy: 0.9248
Epoch 8/20
60000/60000 [==============================] - 3s 45us/sample - loss: 0.2526 - accuracy: 0.9286
Epoch 9/20
60000/60000 [==============================] - 3s 43us/sample - loss: 0.2407 - accuracy: 0.9326
Epoch 10/20
60000/60000 [==============================] - 2s 41us/sample - loss: 0.2299 - accuracy: 0.9350
Epoch 11/20
60000/60000 [==============================] - 2s 40us/sample - loss: 0.2203 - accuracy: 0.9380
Epoch 12/20
60000/60000 [==============================] - 3s 44us/sample - loss: 0.2115 - accuracy: 0.9405
Epoch 13/20
60000/60000 [==============================] - 2s 41us/sample - loss: 0.2033 - accuracy: 0.9426
Epoch 14/20
60000/60000 [==============================] - 2s 39us/sample - loss: 0.1956 - accuracy: 0.9441
Epoch 15/20
60000/60000 [==============================] - 2s 41us/sample - loss: 0.1886 - accuracy: 0.9462
Epoch 16/20
60000/60000 [==============================] - 3s 43us/sample - loss: 0.1816 - accuracy: 0.9485
Epoch 17/20
60000/60000 [==============================] - 2s 40us/sample - loss: 0.1756 - accuracy: 0.9503
Epoch 18/20
60000/60000 [==============================] - 2s 40us/sample - loss: 0.1697 - accuracy: 0.9519
Epoch 19/20
60000/60000 [==============================] - 2s 40us/sample - loss: 0.1644 - accuracy: 0.9538
Epoch 20/20
60000/60000 [==============================] - 2s 39us/sample - loss: 0.1593 - accuracy: 0.9551
Test accuracy: 95.3%