Intermediate Neural Network in Keras

In this notebook, we improve our introductory shallow net from Lesson 1 by applying the theory we have covered since.

Set seed for reproducibility



In [1]:

    
import numpy as np
np.random.seed(42)

Load dependencies



In [2]:

    
import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense
from keras.optimizers import SGD









    



Using TensorFlow backend.

Load data



In [3]:

    
(X_train, y_train), (X_test, y_test) = mnist.load_data()

Preprocess data



In [4]:

    
X_train = X_train.reshape(60000, 784).astype('float32')
X_test = X_test.reshape(10000, 784).astype('float32')



In [5]:

    
X_train /= 255
X_test /= 255



In [6]:

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

Design neural network architecture



In [7]:

    
model = Sequential()
model.add(Dense(64, activation='relu', input_shape=(784,)))
model.add(Dense(64, activation='relu'))
model.add(Dense(10, activation='softmax'))



In [8]:

    
model.summary()









    



_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_1 (Dense)              (None, 64)                50240     
_________________________________________________________________
dense_2 (Dense)              (None, 64)                4160      
_________________________________________________________________
dense_3 (Dense)              (None, 10)                650       
=================================================================
Total params: 55,050
Trainable params: 55,050
Non-trainable params: 0
_________________________________________________________________

Configure model



In [9]:

    
model.compile(loss='categorical_crossentropy', optimizer=SGD(lr=0.1), metrics=['accuracy'])

Train!



In [10]:

    
model.fit(X_train, y_train, batch_size=128, epochs=200, verbose=1, validation_data=(X_test, y_test))









    



Train on 60000 samples, validate on 10000 samples
Epoch 1/200
60000/60000 [==============================] - 1s - loss: 0.4785 - acc: 0.8642 - val_loss: 0.2507 - val_acc: 0.9255
Epoch 2/200
60000/60000 [==============================] - 1s - loss: 0.2245 - acc: 0.9354 - val_loss: 0.1930 - val_acc: 0.9436
Epoch 3/200
60000/60000 [==============================] - 1s - loss: 0.1716 - acc: 0.9500 - val_loss: 0.1506 - val_acc: 0.9547
Epoch 4/200
60000/60000 [==============================] - 1s - loss: 0.1415 - acc: 0.9586 - val_loss: 0.1313 - val_acc: 0.9602
Epoch 5/200
60000/60000 [==============================] - 1s - loss: 0.1201 - acc: 0.9651 - val_loss: 0.1280 - val_acc: 0.9614
Epoch 6/200
60000/60000 [==============================] - 1s - loss: 0.1045 - acc: 0.9697 - val_loss: 0.1061 - val_acc: 0.9669
Epoch 7/200
60000/60000 [==============================] - 1s - loss: 0.0927 - acc: 0.9726 - val_loss: 0.0984 - val_acc: 0.9697
Epoch 8/200
60000/60000 [==============================] - 1s - loss: 0.0826 - acc: 0.9759 - val_loss: 0.0926 - val_acc: 0.9719
Epoch 9/200
60000/60000 [==============================] - 1s - loss: 0.0758 - acc: 0.9774 - val_loss: 0.0904 - val_acc: 0.9732
Epoch 10/200
60000/60000 [==============================] - 1s - loss: 0.0683 - acc: 0.9797 - val_loss: 0.0963 - val_acc: 0.9705
Epoch 11/200
60000/60000 [==============================] - 1s - loss: 0.0631 - acc: 0.9810 - val_loss: 0.0856 - val_acc: 0.9752
Epoch 12/200
60000/60000 [==============================] - 1s - loss: 0.0575 - acc: 0.9832 - val_loss: 0.0839 - val_acc: 0.9749
Epoch 13/200
60000/60000 [==============================] - 1s - loss: 0.0523 - acc: 0.9846 - val_loss: 0.0881 - val_acc: 0.9733
Epoch 14/200
60000/60000 [==============================] - 1s - loss: 0.0488 - acc: 0.9859 - val_loss: 0.0828 - val_acc: 0.9759
Epoch 15/200
60000/60000 [==============================] - 1s - loss: 0.0454 - acc: 0.9869 - val_loss: 0.0844 - val_acc: 0.9735
Epoch 16/200
60000/60000 [==============================] - 1s - loss: 0.0420 - acc: 0.9875 - val_loss: 0.0864 - val_acc: 0.9749
Epoch 17/200
60000/60000 [==============================] - 1s - loss: 0.0400 - acc: 0.9886 - val_loss: 0.0848 - val_acc: 0.9746
Epoch 18/200
60000/60000 [==============================] - 1s - loss: 0.0364 - acc: 0.9894 - val_loss: 0.0748 - val_acc: 0.9775
Epoch 19/200
60000/60000 [==============================] - 1s - loss: 0.0336 - acc: 0.9902 - val_loss: 0.0824 - val_acc: 0.9756
Epoch 20/200
60000/60000 [==============================] - 1s - loss: 0.0315 - acc: 0.9911 - val_loss: 0.0802 - val_acc: 0.9772
Epoch 21/200
60000/60000 [==============================] - 1s - loss: 0.0304 - acc: 0.9914 - val_loss: 0.0791 - val_acc: 0.9759
Epoch 22/200
60000/60000 [==============================] - 1s - loss: 0.0274 - acc: 0.9923 - val_loss: 0.0769 - val_acc: 0.9777
Epoch 23/200
60000/60000 [==============================] - 1s - loss: 0.0255 - acc: 0.9930 - val_loss: 0.0776 - val_acc: 0.9781
Epoch 24/200
60000/60000 [==============================] - 1s - loss: 0.0241 - acc: 0.9937 - val_loss: 0.0783 - val_acc: 0.9771
Epoch 25/200
60000/60000 [==============================] - 1s - loss: 0.0226 - acc: 0.9936 - val_loss: 0.0824 - val_acc: 0.9764
Epoch 26/200
60000/60000 [==============================] - 1s - loss: 0.0212 - acc: 0.9945 - val_loss: 0.0812 - val_acc: 0.9774
Epoch 27/200
60000/60000 [==============================] - 1s - loss: 0.0190 - acc: 0.9954 - val_loss: 0.0795 - val_acc: 0.9784
Epoch 28/200
60000/60000 [==============================] - 1s - loss: 0.0177 - acc: 0.9958 - val_loss: 0.0829 - val_acc: 0.9759
Epoch 29/200
60000/60000 [==============================] - 1s - loss: 0.0166 - acc: 0.9962 - val_loss: 0.0808 - val_acc: 0.9779
Epoch 30/200
60000/60000 [==============================] - 1s - loss: 0.0147 - acc: 0.9970 - val_loss: 0.0836 - val_acc: 0.9774
Epoch 31/200
60000/60000 [==============================] - 1s - loss: 0.0143 - acc: 0.9967 - val_loss: 0.0811 - val_acc: 0.9778
Epoch 32/200
60000/60000 [==============================] - 1s - loss: 0.0127 - acc: 0.9976 - val_loss: 0.0823 - val_acc: 0.9786
Epoch 33/200
60000/60000 [==============================] - 1s - loss: 0.0117 - acc: 0.9977 - val_loss: 0.0843 - val_acc: 0.9772
Epoch 34/200
60000/60000 [==============================] - 1s - loss: 0.0112 - acc: 0.9978 - val_loss: 0.0842 - val_acc: 0.9776
Epoch 35/200
60000/60000 [==============================] - 1s - loss: 0.0104 - acc: 0.9981 - val_loss: 0.0907 - val_acc: 0.9756
Epoch 36/200
60000/60000 [==============================] - 1s - loss: 0.0098 - acc: 0.9981 - val_loss: 0.0853 - val_acc: 0.9775
Epoch 37/200
60000/60000 [==============================] - 1s - loss: 0.0090 - acc: 0.9984 - val_loss: 0.0861 - val_acc: 0.9770
Epoch 38/200
60000/60000 [==============================] - 1s - loss: 0.0081 - acc: 0.9989 - val_loss: 0.0872 - val_acc: 0.9764
Epoch 39/200
60000/60000 [==============================] - 1s - loss: 0.0074 - acc: 0.9991 - val_loss: 0.0918 - val_acc: 0.9768
Epoch 40/200
60000/60000 [==============================] - 1s - loss: 0.0069 - acc: 0.9990 - val_loss: 0.0898 - val_acc: 0.9771
Epoch 41/200
60000/60000 [==============================] - 1s - loss: 0.0068 - acc: 0.9990 - val_loss: 0.0882 - val_acc: 0.9765
Epoch 42/200
60000/60000 [==============================] - 1s - loss: 0.0063 - acc: 0.9993 - val_loss: 0.0909 - val_acc: 0.9765
Epoch 43/200
60000/60000 [==============================] - 1s - loss: 0.0057 - acc: 0.9995 - val_loss: 0.0904 - val_acc: 0.9780
Epoch 44/200
60000/60000 [==============================] - 1s - loss: 0.0051 - acc: 0.9996 - val_loss: 0.0905 - val_acc: 0.9776
Epoch 45/200
60000/60000 [==============================] - 1s - loss: 0.0050 - acc: 0.9996 - val_loss: 0.0917 - val_acc: 0.9773
Epoch 46/200
60000/60000 [==============================] - 1s - loss: 0.0045 - acc: 0.9997 - val_loss: 0.0917 - val_acc: 0.9773
Epoch 47/200
60000/60000 [==============================] - 1s - loss: 0.0043 - acc: 0.9997 - val_loss: 0.0912 - val_acc: 0.9777
Epoch 48/200
60000/60000 [==============================] - 1s - loss: 0.0039 - acc: 0.9998 - val_loss: 0.0943 - val_acc: 0.9769
Epoch 49/200
60000/60000 [==============================] - 1s - loss: 0.0037 - acc: 0.9999 - val_loss: 0.0959 - val_acc: 0.9764
Epoch 50/200
60000/60000 [==============================] - 1s - loss: 0.0036 - acc: 0.9999 - val_loss: 0.0939 - val_acc: 0.9780
Epoch 51/200
60000/60000 [==============================] - 1s - loss: 0.0032 - acc: 0.9999 - val_loss: 0.0928 - val_acc: 0.9774
Epoch 52/200
60000/60000 [==============================] - 1s - loss: 0.0032 - acc: 0.9999 - val_loss: 0.0958 - val_acc: 0.9767
Epoch 53/200
60000/60000 [==============================] - 1s - loss: 0.0031 - acc: 0.9999 - val_loss: 0.0953 - val_acc: 0.9779
Epoch 54/200
60000/60000 [==============================] - 1s - loss: 0.0029 - acc: 0.9999 - val_loss: 0.0965 - val_acc: 0.9768
Epoch 55/200
60000/60000 [==============================] - 1s - loss: 0.0027 - acc: 0.9999 - val_loss: 0.0965 - val_acc: 0.9779
Epoch 56/200
60000/60000 [==============================] - 1s - loss: 0.0026 - acc: 0.9999 - val_loss: 0.0954 - val_acc: 0.9776
Epoch 57/200
60000/60000 [==============================] - 1s - loss: 0.0024 - acc: 0.9999 - val_loss: 0.0961 - val_acc: 0.9781
Epoch 58/200
60000/60000 [==============================] - 1s - loss: 0.0024 - acc: 0.9999 - val_loss: 0.0963 - val_acc: 0.9778
Epoch 59/200
60000/60000 [==============================] - 1s - loss: 0.0023 - acc: 1.0000 - val_loss: 0.0983 - val_acc: 0.9775
Epoch 60/200
60000/60000 [==============================] - 1s - loss: 0.0022 - acc: 1.0000 - val_loss: 0.0989 - val_acc: 0.9776
Epoch 61/200
60000/60000 [==============================] - 1s - loss: 0.0021 - acc: 1.0000 - val_loss: 0.0995 - val_acc: 0.9772
Epoch 62/200
60000/60000 [==============================] - 1s - loss: 0.0021 - acc: 1.0000 - val_loss: 0.1005 - val_acc: 0.9770
Epoch 63/200
60000/60000 [==============================] - 1s - loss: 0.0020 - acc: 1.0000 - val_loss: 0.1007 - val_acc: 0.9772
Epoch 64/200
60000/60000 [==============================] - 1s - loss: 0.0019 - acc: 1.0000 - val_loss: 0.0995 - val_acc: 0.9774
Epoch 65/200
60000/60000 [==============================] - 1s - loss: 0.0019 - acc: 1.0000 - val_loss: 0.1003 - val_acc: 0.9774
Epoch 66/200
60000/60000 [==============================] - 1s - loss: 0.0018 - acc: 1.0000 - val_loss: 0.1015 - val_acc: 0.9777
Epoch 67/200
60000/60000 [==============================] - 1s - loss: 0.0017 - acc: 1.0000 - val_loss: 0.1001 - val_acc: 0.9772
Epoch 68/200
60000/60000 [==============================] - 1s - loss: 0.0017 - acc: 1.0000 - val_loss: 0.1008 - val_acc: 0.9772
Epoch 69/200
60000/60000 [==============================] - 1s - loss: 0.0016 - acc: 1.0000 - val_loss: 0.1028 - val_acc: 0.9772
Epoch 70/200
60000/60000 [==============================] - 1s - loss: 0.0016 - acc: 1.0000 - val_loss: 0.1029 - val_acc: 0.9772
Epoch 71/200
60000/60000 [==============================] - 1s - loss: 0.0016 - acc: 1.0000 - val_loss: 0.1028 - val_acc: 0.9774
Epoch 72/200
60000/60000 [==============================] - 1s - loss: 0.0015 - acc: 1.0000 - val_loss: 0.1032 - val_acc: 0.9773
Epoch 73/200
60000/60000 [==============================] - 1s - loss: 0.0015 - acc: 1.0000 - val_loss: 0.1025 - val_acc: 0.9779
Epoch 74/200
60000/60000 [==============================] - 1s - loss: 0.0015 - acc: 1.0000 - val_loss: 0.1038 - val_acc: 0.9772
Epoch 75/200
60000/60000 [==============================] - 1s - loss: 0.0014 - acc: 1.0000 - val_loss: 0.1037 - val_acc: 0.9773
Epoch 76/200
60000/60000 [==============================] - 1s - loss: 0.0014 - acc: 1.0000 - val_loss: 0.1039 - val_acc: 0.9777
Epoch 77/200
60000/60000 [==============================] - 1s - loss: 0.0014 - acc: 1.0000 - val_loss: 0.1046 - val_acc: 0.9773
Epoch 78/200
60000/60000 [==============================] - 1s - loss: 0.0013 - acc: 1.0000 - val_loss: 0.1044 - val_acc: 0.9773
Epoch 79/200
60000/60000 [==============================] - 1s - loss: 0.0013 - acc: 1.0000 - val_loss: 0.1052 - val_acc: 0.9771
Epoch 80/200
60000/60000 [==============================] - 1s - loss: 0.0013 - acc: 1.0000 - val_loss: 0.1049 - val_acc: 0.9774
Epoch 81/200
60000/60000 [==============================] - 1s - loss: 0.0013 - acc: 1.0000 - val_loss: 0.1077 - val_acc: 0.9770
Epoch 82/200
60000/60000 [==============================] - 1s - loss: 0.0012 - acc: 1.0000 - val_loss: 0.1059 - val_acc: 0.9774
Epoch 83/200
60000/60000 [==============================] - 1s - loss: 0.0012 - acc: 1.0000 - val_loss: 0.1058 - val_acc: 0.9771
Epoch 84/200
60000/60000 [==============================] - 1s - loss: 0.0012 - acc: 1.0000 - val_loss: 0.1064 - val_acc: 0.9773
Epoch 85/200
60000/60000 [==============================] - 1s - loss: 0.0012 - acc: 1.0000 - val_loss: 0.1063 - val_acc: 0.9772
Epoch 86/200
60000/60000 [==============================] - 1s - loss: 0.0012 - acc: 1.0000 - val_loss: 0.1079 - val_acc: 0.9772
Epoch 87/200
60000/60000 [==============================] - 1s - loss: 0.0011 - acc: 1.0000 - val_loss: 0.1066 - val_acc: 0.9774
Epoch 88/200
60000/60000 [==============================] - 1s - loss: 0.0011 - acc: 1.0000 - val_loss: 0.1074 - val_acc: 0.9774
Epoch 89/200
60000/60000 [==============================] - 1s - loss: 0.0011 - acc: 1.0000 - val_loss: 0.1081 - val_acc: 0.9773
Epoch 90/200
60000/60000 [==============================] - 1s - loss: 0.0011 - acc: 1.0000 - val_loss: 0.1079 - val_acc: 0.9775
Epoch 91/200
60000/60000 [==============================] - 1s - loss: 0.0011 - acc: 1.0000 - val_loss: 0.1084 - val_acc: 0.9771
Epoch 92/200
60000/60000 [==============================] - 1s - loss: 0.0010 - acc: 1.0000 - val_loss: 0.1081 - val_acc: 0.9773
Epoch 93/200
60000/60000 [==============================] - 1s - loss: 0.0010 - acc: 1.0000 - val_loss: 0.1087 - val_acc: 0.9773
Epoch 94/200
60000/60000 [==============================] - 1s - loss: 0.0010 - acc: 1.0000 - val_loss: 0.1092 - val_acc: 0.9774
Epoch 95/200
60000/60000 [==============================] - 1s - loss: 9.9927e-04 - acc: 1.0000 - val_loss: 0.1093 - val_acc: 0.9774
Epoch 96/200
60000/60000 [==============================] - 1s - loss: 9.8933e-04 - acc: 1.0000 - val_loss: 0.1092 - val_acc: 0.9772
Epoch 97/200
60000/60000 [==============================] - 1s - loss: 9.7706e-04 - acc: 1.0000 - val_loss: 0.1092 - val_acc: 0.9775
Epoch 98/200
60000/60000 [==============================] - 1s - loss: 9.6611e-04 - acc: 1.0000 - val_loss: 0.1094 - val_acc: 0.9772
Epoch 99/200
60000/60000 [==============================] - 1s - loss: 9.4689e-04 - acc: 1.0000 - val_loss: 0.1097 - val_acc: 0.9775
Epoch 100/200
60000/60000 [==============================] - 1s - loss: 9.3695e-04 - acc: 1.0000 - val_loss: 0.1095 - val_acc: 0.9774
Epoch 101/200
60000/60000 [==============================] - 1s - loss: 9.3021e-04 - acc: 1.0000 - val_loss: 0.1105 - val_acc: 0.9775
Epoch 102/200
60000/60000 [==============================] - 1s - loss: 9.1504e-04 - acc: 1.0000 - val_loss: 0.1107 - val_acc: 0.9773
Epoch 103/200
60000/60000 [==============================] - 1s - loss: 9.1007e-04 - acc: 1.0000 - val_loss: 0.1109 - val_acc: 0.9775
Epoch 104/200
60000/60000 [==============================] - 1s - loss: 8.9332e-04 - acc: 1.0000 - val_loss: 0.1107 - val_acc: 0.9773
Epoch 105/200
60000/60000 [==============================] - 1s - loss: 8.8180e-04 - acc: 1.0000 - val_loss: 0.1111 - val_acc: 0.9774
Epoch 106/200
60000/60000 [==============================] - 1s - loss: 8.7657e-04 - acc: 1.0000 - val_loss: 0.1117 - val_acc: 0.9773
Epoch 107/200
60000/60000 [==============================] - 1s - loss: 8.6383e-04 - acc: 1.0000 - val_loss: 0.1117 - val_acc: 0.9774
Epoch 108/200
60000/60000 [==============================] - 1s - loss: 8.5332e-04 - acc: 1.0000 - val_loss: 0.1114 - val_acc: 0.9777
Epoch 109/200
60000/60000 [==============================] - 1s - loss: 8.4448e-04 - acc: 1.0000 - val_loss: 0.1121 - val_acc: 0.9773
Epoch 110/200
60000/60000 [==============================] - 1s - loss: 8.3663e-04 - acc: 1.0000 - val_loss: 0.1123 - val_acc: 0.9776
Epoch 111/200
60000/60000 [==============================] - 1s - loss: 8.2692e-04 - acc: 1.0000 - val_loss: 0.1123 - val_acc: 0.9774
Epoch 112/200
60000/60000 [==============================] - 1s - loss: 8.1818e-04 - acc: 1.0000 - val_loss: 0.1126 - val_acc: 0.9772
Epoch 113/200
60000/60000 [==============================] - 1s - loss: 8.1627e-04 - acc: 1.0000 - val_loss: 0.1124 - val_acc: 0.9774
Epoch 114/200
60000/60000 [==============================] - 1s - loss: 8.0304e-04 - acc: 1.0000 - val_loss: 0.1128 - val_acc: 0.9772
Epoch 115/200
60000/60000 [==============================] - 1s - loss: 7.9308e-04 - acc: 1.0000 - val_loss: 0.1130 - val_acc: 0.9775
Epoch 116/200
60000/60000 [==============================] - 1s - loss: 7.8779e-04 - acc: 1.0000 - val_loss: 0.1130 - val_acc: 0.9771
Epoch 117/200
60000/60000 [==============================] - 1s - loss: 7.8072e-04 - acc: 1.0000 - val_loss: 0.1140 - val_acc: 0.9774
Epoch 118/200
60000/60000 [==============================] - 1s - loss: 7.6669e-04 - acc: 1.0000 - val_loss: 0.1136 - val_acc: 0.9775
Epoch 119/200
60000/60000 [==============================] - 1s - loss: 7.6492e-04 - acc: 1.0000 - val_loss: 0.1138 - val_acc: 0.9772
Epoch 120/200
60000/60000 [==============================] - 1s - loss: 7.5824e-04 - acc: 1.0000 - val_loss: 0.1138 - val_acc: 0.9775
Epoch 121/200
60000/60000 [==============================] - 1s - loss: 7.5277e-04 - acc: 1.0000 - val_loss: 0.1142 - val_acc: 0.9769
Epoch 122/200
60000/60000 [==============================] - 1s - loss: 7.4638e-04 - acc: 1.0000 - val_loss: 0.1144 - val_acc: 0.9775
Epoch 123/200
60000/60000 [==============================] - 1s - loss: 7.3855e-04 - acc: 1.0000 - val_loss: 0.1142 - val_acc: 0.9773
Epoch 124/200
60000/60000 [==============================] - 1s - loss: 7.3353e-04 - acc: 1.0000 - val_loss: 0.1144 - val_acc: 0.9775
Epoch 125/200
60000/60000 [==============================] - 1s - loss: 7.2691e-04 - acc: 1.0000 - val_loss: 0.1147 - val_acc: 0.9774
Epoch 126/200
60000/60000 [==============================] - 1s - loss: 7.1877e-04 - acc: 1.0000 - val_loss: 0.1147 - val_acc: 0.9775
Epoch 127/200
60000/60000 [==============================] - 1s - loss: 7.1330e-04 - acc: 1.0000 - val_loss: 0.1148 - val_acc: 0.9771
Epoch 128/200
60000/60000 [==============================] - 1s - loss: 7.0961e-04 - acc: 1.0000 - val_loss: 0.1154 - val_acc: 0.9774
Epoch 129/200
60000/60000 [==============================] - 1s - loss: 7.0377e-04 - acc: 1.0000 - val_loss: 0.1147 - val_acc: 0.9774
Epoch 130/200
60000/60000 [==============================] - 1s - loss: 6.9733e-04 - acc: 1.0000 - val_loss: 0.1154 - val_acc: 0.9773
Epoch 131/200
60000/60000 [==============================] - 1s - loss: 6.9228e-04 - acc: 1.0000 - val_loss: 0.1156 - val_acc: 0.9772
Epoch 132/200
60000/60000 [==============================] - 1s - loss: 6.8761e-04 - acc: 1.0000 - val_loss: 0.1159 - val_acc: 0.9771
Epoch 133/200
60000/60000 [==============================] - 1s - loss: 6.8353e-04 - acc: 1.0000 - val_loss: 0.1158 - val_acc: 0.9774
Epoch 134/200
60000/60000 [==============================] - 1s - loss: 6.7750e-04 - acc: 1.0000 - val_loss: 0.1162 - val_acc: 0.9771
Epoch 135/200
60000/60000 [==============================] - 1s - loss: 6.7072e-04 - acc: 1.0000 - val_loss: 0.1158 - val_acc: 0.9775
Epoch 136/200
60000/60000 [==============================] - 1s - loss: 6.6779e-04 - acc: 1.0000 - val_loss: 0.1166 - val_acc: 0.9774
Epoch 137/200
60000/60000 [==============================] - 1s - loss: 6.6279e-04 - acc: 1.0000 - val_loss: 0.1166 - val_acc: 0.9772
Epoch 138/200
60000/60000 [==============================] - 1s - loss: 6.5824e-04 - acc: 1.0000 - val_loss: 0.1171 - val_acc: 0.9773
Epoch 139/200
60000/60000 [==============================] - 1s - loss: 6.5465e-04 - acc: 1.0000 - val_loss: 0.1173 - val_acc: 0.9773
Epoch 140/200
60000/60000 [==============================] - 1s - loss: 6.5011e-04 - acc: 1.0000 - val_loss: 0.1170 - val_acc: 0.9772
Epoch 141/200
60000/60000 [==============================] - 1s - loss: 6.4690e-04 - acc: 1.0000 - val_loss: 0.1177 - val_acc: 0.9773
Epoch 142/200
60000/60000 [==============================] - 1s - loss: 6.4178e-04 - acc: 1.0000 - val_loss: 0.1174 - val_acc: 0.9773
Epoch 143/200
60000/60000 [==============================] - 1s - loss: 6.3734e-04 - acc: 1.0000 - val_loss: 0.1175 - val_acc: 0.9773
Epoch 144/200
60000/60000 [==============================] - 1s - loss: 6.3393e-04 - acc: 1.0000 - val_loss: 0.1177 - val_acc: 0.9773
Epoch 145/200
60000/60000 [==============================] - 1s - loss: 6.2917e-04 - acc: 1.0000 - val_loss: 0.1179 - val_acc: 0.9775
Epoch 146/200
60000/60000 [==============================] - 1s - loss: 6.2485e-04 - acc: 1.0000 - val_loss: 0.1182 - val_acc: 0.9771
Epoch 147/200
60000/60000 [==============================] - 1s - loss: 6.2252e-04 - acc: 1.0000 - val_loss: 0.1175 - val_acc: 0.9772
Epoch 148/200
60000/60000 [==============================] - 1s - loss: 6.1810e-04 - acc: 1.0000 - val_loss: 0.1177 - val_acc: 0.9772
Epoch 149/200
60000/60000 [==============================] - 1s - loss: 6.1423e-04 - acc: 1.0000 - val_loss: 0.1187 - val_acc: 0.9774
Epoch 150/200
60000/60000 [==============================] - 1s - loss: 6.0999e-04 - acc: 1.0000 - val_loss: 0.1187 - val_acc: 0.9774
Epoch 151/200
60000/60000 [==============================] - 1s - loss: 6.0768e-04 - acc: 1.0000 - val_loss: 0.1187 - val_acc: 0.9776
Epoch 152/200
60000/60000 [==============================] - 1s - loss: 6.0392e-04 - acc: 1.0000 - val_loss: 0.1186 - val_acc: 0.9773
Epoch 153/200
60000/60000 [==============================] - 1s - loss: 6.0023e-04 - acc: 1.0000 - val_loss: 0.1188 - val_acc: 0.9774
Epoch 154/200
60000/60000 [==============================] - 1s - loss: 5.9694e-04 - acc: 1.0000 - val_loss: 0.1193 - val_acc: 0.9772
Epoch 155/200
60000/60000 [==============================] - 1s - loss: 5.9343e-04 - acc: 1.0000 - val_loss: 0.1195 - val_acc: 0.9771
Epoch 156/200
60000/60000 [==============================] - 2s - loss: 5.9063e-04 - acc: 1.0000 - val_loss: 0.1192 - val_acc: 0.9772
Epoch 157/200
60000/60000 [==============================] - 2s - loss: 5.8791e-04 - acc: 1.0000 - val_loss: 0.1193 - val_acc: 0.9771
Epoch 158/200
60000/60000 [==============================] - 2s - loss: 5.8399e-04 - acc: 1.0000 - val_loss: 0.1194 - val_acc: 0.9773
Epoch 159/200
60000/60000 [==============================] - 2s - loss: 5.8049e-04 - acc: 1.0000 - val_loss: 0.1195 - val_acc: 0.9774
Epoch 160/200
60000/60000 [==============================] - 1s - loss: 5.7753e-04 - acc: 1.0000 - val_loss: 0.1192 - val_acc: 0.9775
Epoch 161/200
60000/60000 [==============================] - 1s - loss: 5.7520e-04 - acc: 1.0000 - val_loss: 0.1197 - val_acc: 0.9774
Epoch 162/200
60000/60000 [==============================] - 1s - loss: 5.7192e-04 - acc: 1.0000 - val_loss: 0.1197 - val_acc: 0.9771
Epoch 163/200
60000/60000 [==============================] - 1s - loss: 5.6971e-04 - acc: 1.0000 - val_loss: 0.1204 - val_acc: 0.9773
Epoch 164/200
60000/60000 [==============================] - 1s - loss: 5.6689e-04 - acc: 1.0000 - val_loss: 0.1201 - val_acc: 0.9773
Epoch 165/200
60000/60000 [==============================] - 1s - loss: 5.6463e-04 - acc: 1.0000 - val_loss: 0.1202 - val_acc: 0.9771
Epoch 166/200
60000/60000 [==============================] - 1s - loss: 5.6155e-04 - acc: 1.0000 - val_loss: 0.1206 - val_acc: 0.9773
Epoch 167/200
60000/60000 [==============================] - 1s - loss: 5.5831e-04 - acc: 1.0000 - val_loss: 0.1208 - val_acc: 0.9772
Epoch 168/200
60000/60000 [==============================] - 1s - loss: 5.5547e-04 - acc: 1.0000 - val_loss: 0.1207 - val_acc: 0.9770
Epoch 169/200
60000/60000 [==============================] - 1s - loss: 5.5346e-04 - acc: 1.0000 - val_loss: 0.1210 - val_acc: 0.9772
Epoch 170/200
60000/60000 [==============================] - 1s - loss: 5.5140e-04 - acc: 1.0000 - val_loss: 0.1208 - val_acc: 0.9771
Epoch 171/200
60000/60000 [==============================] - 1s - loss: 5.4791e-04 - acc: 1.0000 - val_loss: 0.1207 - val_acc: 0.9771
Epoch 172/200
60000/60000 [==============================] - 1s - loss: 5.4535e-04 - acc: 1.0000 - val_loss: 0.1211 - val_acc: 0.9774
Epoch 173/200
60000/60000 [==============================] - 1s - loss: 5.4422e-04 - acc: 1.0000 - val_loss: 0.1216 - val_acc: 0.9771
Epoch 174/200
60000/60000 [==============================] - 1s - loss: 5.4011e-04 - acc: 1.0000 - val_loss: 0.1210 - val_acc: 0.9773
Epoch 175/200
60000/60000 [==============================] - 1s - loss: 5.3945e-04 - acc: 1.0000 - val_loss: 0.1213 - val_acc: 0.9773
Epoch 176/200
60000/60000 [==============================] - 1s - loss: 5.3635e-04 - acc: 1.0000 - val_loss: 0.1212 - val_acc: 0.9773
Epoch 177/200
60000/60000 [==============================] - 1s - loss: 5.3404e-04 - acc: 1.0000 - val_loss: 0.1217 - val_acc: 0.9773
Epoch 178/200
60000/60000 [==============================] - 1s - loss: 5.3159e-04 - acc: 1.0000 - val_loss: 0.1216 - val_acc: 0.9773
Epoch 179/200
60000/60000 [==============================] - 1s - loss: 5.2900e-04 - acc: 1.0000 - val_loss: 0.1216 - val_acc: 0.9772
Epoch 180/200
60000/60000 [==============================] - 1s - loss: 5.2754e-04 - acc: 1.0000 - val_loss: 0.1216 - val_acc: 0.9774
Epoch 181/200
60000/60000 [==============================] - 1s - loss: 5.2574e-04 - acc: 1.0000 - val_loss: 0.1221 - val_acc: 0.9775
Epoch 182/200
60000/60000 [==============================] - 1s - loss: 5.2277e-04 - acc: 1.0000 - val_loss: 0.1223 - val_acc: 0.9770
Epoch 183/200
60000/60000 [==============================] - 1s - loss: 5.2182e-04 - acc: 1.0000 - val_loss: 0.1225 - val_acc: 0.9773
Epoch 184/200
60000/60000 [==============================] - 1s - loss: 5.1933e-04 - acc: 1.0000 - val_loss: 0.1228 - val_acc: 0.9772
Epoch 185/200
60000/60000 [==============================] - 1s - loss: 5.1725e-04 - acc: 1.0000 - val_loss: 0.1226 - val_acc: 0.9774
Epoch 186/200
60000/60000 [==============================] - 1s - loss: 5.1478e-04 - acc: 1.0000 - val_loss: 0.1225 - val_acc: 0.9773
Epoch 187/200
60000/60000 [==============================] - 1s - loss: 5.1329e-04 - acc: 1.0000 - val_loss: 0.1227 - val_acc: 0.9774
Epoch 188/200
60000/60000 [==============================] - 1s - loss: 5.1120e-04 - acc: 1.0000 - val_loss: 0.1227 - val_acc: 0.9772
Epoch 189/200
60000/60000 [==============================] - 1s - loss: 5.0978e-04 - acc: 1.0000 - val_loss: 0.1231 - val_acc: 0.9772
Epoch 190/200
60000/60000 [==============================] - 1s - loss: 5.0732e-04 - acc: 1.0000 - val_loss: 0.1230 - val_acc: 0.9773
Epoch 191/200
60000/60000 [==============================] - 1s - loss: 5.0517e-04 - acc: 1.0000 - val_loss: 0.1228 - val_acc: 0.9774
Epoch 192/200
60000/60000 [==============================] - 1s - loss: 5.0432e-04 - acc: 1.0000 - val_loss: 0.1233 - val_acc: 0.9774
Epoch 193/200
60000/60000 [==============================] - 1s - loss: 5.0206e-04 - acc: 1.0000 - val_loss: 0.1232 - val_acc: 0.9774
Epoch 194/200
60000/60000 [==============================] - 1s - loss: 5.0064e-04 - acc: 1.0000 - val_loss: 0.1235 - val_acc: 0.9772
Epoch 195/200
60000/60000 [==============================] - 1s - loss: 4.9799e-04 - acc: 1.0000 - val_loss: 0.1237 - val_acc: 0.9774
Epoch 196/200
60000/60000 [==============================] - 1s - loss: 4.9632e-04 - acc: 1.0000 - val_loss: 0.1237 - val_acc: 0.9772
Epoch 197/200
60000/60000 [==============================] - 1s - loss: 4.9491e-04 - acc: 1.0000 - val_loss: 0.1240 - val_acc: 0.9774
Epoch 198/200
60000/60000 [==============================] - 1s - loss: 4.9344e-04 - acc: 1.0000 - val_loss: 0.1242 - val_acc: 0.9772
Epoch 199/200
60000/60000 [==============================] - 1s - loss: 4.9187e-04 - acc: 1.0000 - val_loss: 0.1239 - val_acc: 0.9775
Epoch 200/200
60000/60000 [==============================] - 1s - loss: 4.8932e-04 - acc: 1.0000 - val_loss: 0.1241 - val_acc: 0.9774






    Out[10]:





<keras.callbacks.History at 0x7f3096adadd8>



In [ ]: