#This is a sample Keras code to predict MNIST Digits using simple Multi-Layer Perceptron#Load the MNIST Dataset from Keras
from keras.datasets import mnistUsing TensorFlow backend.
(x_train,y_train),(x_test,y_test) = mnist.load_data()#show the dimensions of the loaded training and test data sets
x_train.shape
#x_train is an array of gray-scale images (28x28 pixels)(60000, 28, 28)
y_train.shape
#y_train is an array of labels corresponding to the x_train images(60000,)
x_test.shape
#x_test is an array of gray-scale images that will be used for testing(10000, 28, 28)
y_test.shape
#y_test is an array of labels for testing(10000,)
#display some sample images from the training and testing datasets
#show first image
%matplotlib inline
import matplotlib.pyplot as plt
img = x_train[0]
plt.imshow(img, cmap='gray')<matplotlib.image.AxesImage at 0x118dbd550>
img = x_test[0]
plt.imshow(img, cmap='gray')<matplotlib.image.AxesImage at 0x118eab550>
#show the corresponding labels
label = y_train[0]
print(label)
label = y_test[0]
print(label)5
7
dim = img.shape[0]
#flatten the images
flat_train = x_train.reshape([-1,dim*dim])
flat_test = x_test.reshape([-1,dim*dim])
print('flattened shape: ', flat_train.shape)flattened shape: (60000, 784)
#show sample values
flat_train[0]array([ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 18, 18, 18,
126, 136, 175, 26, 166, 255, 247, 127, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 30, 36, 94, 154, 170, 253,
253, 253, 253, 253, 225, 172, 253, 242, 195, 64, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 49, 238, 253, 253, 253,
253, 253, 253, 253, 253, 251, 93, 82, 82, 56, 39, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 18, 219, 253,
253, 253, 253, 253, 198, 182, 247, 241, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
80, 156, 107, 253, 253, 205, 11, 0, 43, 154, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 14, 1, 154, 253, 90, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 139, 253, 190, 2, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 11, 190, 253, 70,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 35,
241, 225, 160, 108, 1, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 81, 240, 253, 253, 119, 25, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 45, 186, 253, 253, 150, 27, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 16, 93, 252, 253, 187,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 249,
253, 249, 64, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 46, 130,
183, 253, 253, 207, 2, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 39, 148,
229, 253, 253, 253, 250, 182, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 24, 114,
221, 253, 253, 253, 253, 201, 78, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 23, 66,
213, 253, 253, 253, 253, 198, 81, 2, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 18, 171,
219, 253, 253, 253, 253, 195, 80, 9, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 55, 172,
226, 253, 253, 253, 253, 244, 133, 11, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
136, 253, 253, 253, 212, 135, 132, 16, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0], dtype=uint8)
#normalize the values
norm_train = flat_train.astype('float32')/255
norm_test = flat_test.astype('float32')/255#show sample normalized values
norm_train[0]array([ 0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0.01176471, 0.07058824, 0.07058824,
0.07058824, 0.49411765, 0.53333336, 0.68627453, 0.10196079,
0.65098041, 1. , 0.96862745, 0.49803922, 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0.11764706, 0.14117648, 0.36862746, 0.60392159,
0.66666669, 0.99215686, 0.99215686, 0.99215686, 0.99215686,
0.99215686, 0.88235295, 0.67450982, 0.99215686, 0.94901961,
0.7647059 , 0.25098041, 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0.19215687, 0.93333334,
0.99215686, 0.99215686, 0.99215686, 0.99215686, 0.99215686,
0.99215686, 0.99215686, 0.99215686, 0.98431373, 0.36470589,
0.32156864, 0.32156864, 0.21960784, 0.15294118, 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0.07058824, 0.85882354, 0.99215686, 0.99215686,
0.99215686, 0.99215686, 0.99215686, 0.7764706 , 0.71372551,
0.96862745, 0.94509804, 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0.3137255 , 0.61176473, 0.41960785, 0.99215686, 0.99215686,
0.80392158, 0.04313726, 0. , 0.16862746, 0.60392159,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0.05490196,
0.00392157, 0.60392159, 0.99215686, 0.35294119, 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0.54509807,
0.99215686, 0.74509805, 0.00784314, 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0.04313726, 0.74509805, 0.99215686,
0.27450982, 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0.13725491, 0.94509804, 0.88235295, 0.627451 ,
0.42352942, 0.00392157, 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0.31764707, 0.94117647, 0.99215686, 0.99215686, 0.46666667,
0.09803922, 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0.17647059,
0.72941178, 0.99215686, 0.99215686, 0.58823532, 0.10588235,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0.0627451 , 0.36470589,
0.98823529, 0.99215686, 0.73333335, 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0.97647059, 0.99215686,
0.97647059, 0.25098041, 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0.18039216, 0.50980395,
0.71764708, 0.99215686, 0.99215686, 0.81176472, 0.00784314,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0.15294118,
0.58039218, 0.89803922, 0.99215686, 0.99215686, 0.99215686,
0.98039216, 0.71372551, 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0.09411765, 0.44705883, 0.86666667, 0.99215686, 0.99215686,
0.99215686, 0.99215686, 0.78823531, 0.30588236, 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0.09019608, 0.25882354, 0.83529413, 0.99215686,
0.99215686, 0.99215686, 0.99215686, 0.7764706 , 0.31764707,
0.00784314, 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0.07058824, 0.67058825, 0.85882354,
0.99215686, 0.99215686, 0.99215686, 0.99215686, 0.7647059 ,
0.3137255 , 0.03529412, 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0.21568628, 0.67450982,
0.88627452, 0.99215686, 0.99215686, 0.99215686, 0.99215686,
0.95686275, 0.52156866, 0.04313726, 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0.53333336, 0.99215686, 0.99215686, 0.99215686,
0.83137256, 0.52941179, 0.51764709, 0.0627451 , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. , 0. ,
0. , 0. , 0. , 0. ], dtype=float32)
#transform labels as categorical values
from keras.utils import to_categorical
train_labels = to_categorical(y_train)
test_labels = to_categorical(y_test)
print('label shape: ', train_labels.shape)label shape: (60000, 10)
#set configurable parameters
input_size = norm_train.shape[1]
num_labels = train_labels.shape[1]
batch_size = 128
hidden_units = 256
dropout = 0.45from keras.models import Sequential
from keras.layers import Dense, Activation, Dropout
model = Sequential()
model.add(Dense(hidden_units, input_dim=input_size))
model.add(Activation('relu'))
model.add(Dropout(dropout))
model.add(Dense(hidden_units))
model.add(Activation('relu'))
model.add(Dropout(dropout))
model.add(Dense(num_labels))
model.add(Activation('softmax'))
model.summary()_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
dense_3 (Dense) (None, 256) 200960
_________________________________________________________________
activation_3 (Activation) (None, 256) 0
_________________________________________________________________
dropout_3 (Dropout) (None, 256) 0
_________________________________________________________________
dense_4 (Dense) (None, 256) 65792
_________________________________________________________________
activation_4 (Activation) (None, 256) 0
_________________________________________________________________
dropout_4 (Dropout) (None, 256) 0
_________________________________________________________________
dense_5 (Dense) (None, 10) 2570
_________________________________________________________________
activation_5 (Activation) (None, 10) 0
=================================================================
Total params: 269,322
Trainable params: 269,322
Non-trainable params: 0
_________________________________________________________________
#compile our model
model.compile(loss='categorical_crossentropy', optimizer='adam',
metrics=['accuracy'])#train our model
model.fit(norm_train, train_labels, epochs=20, batch_size=batch_size)Epoch 1/20
60000/60000 [==============================] - 8s 136us/step - loss: 0.4303 - acc: 0.8659
Epoch 2/20
60000/60000 [==============================] - 8s 133us/step - loss: 0.2007 - acc: 0.9402
Epoch 3/20
60000/60000 [==============================] - 8s 133us/step - loss: 0.1536 - acc: 0.9541
Epoch 4/20
60000/60000 [==============================] - 8s 128us/step - loss: 0.1300 - acc: 0.9610
Epoch 5/20
60000/60000 [==============================] - 7s 121us/step - loss: 0.1153 - acc: 0.9652
Epoch 6/20
60000/60000 [==============================] - 7s 120us/step - loss: 0.1027 - acc: 0.9692
Epoch 7/20
60000/60000 [==============================] - 7s 120us/step - loss: 0.0958 - acc: 0.9708
Epoch 8/20
60000/60000 [==============================] - 7s 120us/step - loss: 0.0886 - acc: 0.9724
Epoch 9/20
60000/60000 [==============================] - 7s 120us/step - loss: 0.0831 - acc: 0.9744
Epoch 10/20
60000/60000 [==============================] - 7s 120us/step - loss: 0.0765 - acc: 0.9763
Epoch 11/20
60000/60000 [==============================] - 7s 121us/step - loss: 0.0736 - acc: 0.9763
Epoch 12/20
60000/60000 [==============================] - 8s 141us/step - loss: 0.0709 - acc: 0.9780
Epoch 13/20
60000/60000 [==============================] - 8s 137us/step - loss: 0.0664 - acc: 0.9788
Epoch 14/20
60000/60000 [==============================] - 8s 138us/step - loss: 0.0638 - acc: 0.9793
Epoch 15/20
60000/60000 [==============================] - 8s 137us/step - loss: 0.0644 - acc: 0.9793
Epoch 16/20
60000/60000 [==============================] - 8s 138us/step - loss: 0.0579 - acc: 0.9809
Epoch 17/20
60000/60000 [==============================] - 8s 140us/step - loss: 0.0593 - acc: 0.9812
Epoch 18/20
60000/60000 [==============================] - 8s 139us/step - loss: 0.0550 - acc: 0.9814
Epoch 19/20
60000/60000 [==============================] - 8s 138us/step - loss: 0.0550 - acc: 0.9824
Epoch 20/20
60000/60000 [==============================] - 8s 139us/step - loss: 0.0566 - acc: 0.9818
<keras.callbacks.History at 0x126ce9b00>
score = model.evaluate(norm_test, test_labels, batch_size=batch_size)10000/10000 [==============================] - 0s 45us/step
score[0.058491837208101062, 0.98309999999999997]
sample = norm_train[0]
sample = sample.reshape([1,-1])num = model.predict(sample)numarray([[ 1.70953655e-21, 7.53669376e-16, 6.53810571e-17,
3.69392546e-05, 8.36950003e-20, 9.99962926e-01,
6.04128879e-19, 7.87650975e-17, 1.46094193e-14,
1.53700356e-07]], dtype=float32)
num.argmax()5
sample = norm_test[0].reshape([1,-1])
num = model.predict(sample)
print("probability: ", num)
print("prediction: ", num.argmax())probability: [[ 1.49344201e-13 1.42855963e-10 1.17309973e-09 1.21340813e-08
2.70941496e-14 1.09324581e-11 6.53023732e-19 9.99999523e-01
4.81604201e-11 4.57720859e-07]]
prediction: 7