Deep Learning Hello World Part5
Deep Learning Hello World! (LeNet-5)
Objective: To be able to implement LeNet-5 for MNIST Classification
Step 1: Taking care of the imports which includes numpy, datasets, models, layers, optimizers, and utils.
You will also be able to tell if your set-up is correct/complete.
from keras import backend as K
from keras.models import Sequential
from keras.models import model_from_json
from keras.layers.convolutional import Conv2D
from keras.layers.convolutional import MaxPooling2D
from keras.layers.core import Activation
from keras.layers.core import Flatten
from keras.layers.core import Dense
from keras.datasets import mnist
from keras.utils import np_utils
from keras.optimizers import Adam
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
from IPython.display import Image # for displaying images
from IPython.core.display import HTML
Step 2: Define LeNet-5 CNN model
Image(url= "https://world4jason.gitbooks.io/research-log/content/deepLearning/CNN/img/lenet.png")
class LeNet:
@staticmethod
def build(input_shape, classes):
model = Sequential()
# CONV => RELU => POOL
model.add(Conv2D(20, kernel_size=5, padding="same",
input_shape=input_shape))
model.add(Activation("relu"))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
# CONV => RELU => POOL
model.add(Conv2D(50, kernel_size=5, padding="same"))
model.add(Activation("relu"))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
# Flatten => RELU layers
model.add(Flatten())
model.add(Dense(500))
model.add(Activation("relu"))
# a softmax classifier
model.add(Dense(classes))
model.add(Activation("softmax"))
return model
Step 3: Set-up some constants to be utilized in the training/testing of the model <br>
NB_EPOCH = 250
BATCH_SIZE = 128
VERBOSE = 1
OPTIMIZER = Adam()
VALIDATION_SPLIT=0.2
IMG_ROWS, IMG_COLS = 28, 28
NB_CLASSES = 10
INPUT_SHAPE = (1, IMG_ROWS, IMG_COLS)
np.random.seed(1983) # for reproducibility
Step 4: Load the MNIST Dataset which are shuffled and split between train and test sets <br>
- X_train is 60000 rows of 28x28 values
- X_test is 10000 rows of 28x28 values
(X_train, y_train), (X_test, y_test) = mnist.load_data()
K.set_image_dim_ordering("th")
print("First 100 train images:")
for k in range(100):
plt.subplot(10, 10, k+1)
plt.gca().axes.get_yaxis().set_visible(False)
plt.gca().axes.get_xaxis().set_visible(False)
plt.imshow(X_train[k])
First 100 train images:
Step 5: Preprocess the input data by reshaping it, converting it to float, and normalizing it [0-1].
X_train = X_train.astype('float32')
X_test = X_test.astype('float32')
X_train /= 255
X_test /= 255
# we need a 60K x [1 x 28 x 28] shape as input to the CONVNET
X_train = X_train[:, np.newaxis, :, :]
X_test = X_test[:, np.newaxis, :, :]
print(X_train.shape, 'train samples')
print(X_test.shape, 'test samples')
(60000, 1, 28, 28) train samples
(10000, 1, 28, 28) test samples
Step 6: Convert class vectors to binary class matrices; One-Hot-Encoding (OHE)
Y_train = np_utils.to_categorical(y_train, NB_CLASSES)
Y_test = np_utils.to_categorical(y_test, NB_CLASSES)
Step 7: Create the model with 3 layers: Input:784 ==> Hidden:128 w/ dropout ==> Hidden:128 w/ dropout ==> Output:10 (with Softmax activation)
model = LeNet.build(input_shape=INPUT_SHAPE, classes=NB_CLASSES)
model.summary()
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
conv2d_1 (Conv2D) (None, 20, 28, 28) 520
_________________________________________________________________
activation_1 (Activation) (None, 20, 28, 28) 0
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 20, 14, 14) 0
_________________________________________________________________
conv2d_2 (Conv2D) (None, 50, 14, 14) 25050
_________________________________________________________________
activation_2 (Activation) (None, 50, 14, 14) 0
_________________________________________________________________
max_pooling2d_2 (MaxPooling2 (None, 50, 7, 7) 0
_________________________________________________________________
flatten_1 (Flatten) (None, 2450) 0
_________________________________________________________________
dense_1 (Dense) (None, 500) 1225500
_________________________________________________________________
activation_3 (Activation) (None, 500) 0
_________________________________________________________________
dense_2 (Dense) (None, 10) 5010
_________________________________________________________________
activation_4 (Activation) (None, 10) 0
=================================================================
Total params: 1,256,080
Trainable params: 1,256,080
Non-trainable params: 0
_________________________________________________________________
Step 8: Compile the model with categorical_crossentropy loss function, Adam optimizer, and accuracy metric
model.compile(loss='categorical_crossentropy',
optimizer=OPTIMIZER,
metrics=['accuracy'])
Step 9: Perform the training with 128 batch size, 250 epochs, and 20 % of the train data used for validation
history = model.fit(X_train, Y_train,
batch_size=BATCH_SIZE, epochs=NB_EPOCH,
verbose=VERBOSE, validation_split=VALIDATION_SPLIT)
Train on 48000 samples, validate on 12000 samples
Epoch 1/250
48000/48000 [==============================] - 6s - loss: 0.0030 - acc: 0.9990 - val_loss: 0.0417 - val_acc: 0.9908
Epoch 2/250
48000/48000 [==============================] - 6s - loss: 0.0021 - acc: 0.9994 - val_loss: 0.0736 - val_acc: 0.9873
Epoch 3/250
48000/48000 [==============================] - 6s - loss: 0.0018 - acc: 0.9995 - val_loss: 0.0454 - val_acc: 0.9916
Epoch 4/250
48000/48000 [==============================] - 6s - loss: 0.0043 - acc: 0.9987 - val_loss: 0.0511 - val_acc: 0.9900
Epoch 5/250
48000/48000 [==============================] - 6s - loss: 0.0033 - acc: 0.9989 - val_loss: 0.0636 - val_acc: 0.9893
Epoch 6/250
48000/48000 [==============================] - 6s - loss: 0.0040 - acc: 0.9988 - val_loss: 0.0474 - val_acc: 0.9910
Epoch 7/250
48000/48000 [==============================] - 6s - loss: 9.1074e-04 - acc: 0.9998 - val_loss: 0.0420 - val_acc: 0.9930
Epoch 8/250
48000/48000 [==============================] - 6s - loss: 3.2303e-05 - acc: 1.0000 - val_loss: 0.0431 - val_acc: 0.9927
Epoch 9/250
48000/48000 [==============================] - 6s - loss: 8.9252e-06 - acc: 1.0000 - val_loss: 0.0429 - val_acc: 0.9934
Epoch 10/250
48000/48000 [==============================] - 6s - loss: 6.4166e-06 - acc: 1.0000 - val_loss: 0.0432 - val_acc: 0.9933
Epoch 11/250
48000/48000 [==============================] - 6s - loss: 4.9292e-06 - acc: 1.0000 - val_loss: 0.0436 - val_acc: 0.9935
Epoch 12/250
48000/48000 [==============================] - 6s - loss: 3.9686e-06 - acc: 1.0000 - val_loss: 0.0439 - val_acc: 0.9935
Epoch 13/250
48000/48000 [==============================] - 6s - loss: 3.2273e-06 - acc: 1.0000 - val_loss: 0.0442 - val_acc: 0.9935
Epoch 14/250
48000/48000 [==============================] - 6s - loss: 2.6348e-06 - acc: 1.0000 - val_loss: 0.0445 - val_acc: 0.9935
Epoch 15/250
48000/48000 [==============================] - 6s - loss: 2.1793e-06 - acc: 1.0000 - val_loss: 0.0449 - val_acc: 0.9935
Epoch 16/250
48000/48000 [==============================] - 6s - loss: 1.8254e-06 - acc: 1.0000 - val_loss: 0.0452 - val_acc: 0.9936
Epoch 17/250
48000/48000 [==============================] - 6s - loss: 1.5246e-06 - acc: 1.0000 - val_loss: 0.0456 - val_acc: 0.9936
Epoch 18/250
48000/48000 [==============================] - 6s - loss: 1.2681e-06 - acc: 1.0000 - val_loss: 0.0459 - val_acc: 0.9936
Epoch 19/250
48000/48000 [==============================] - 6s - loss: 1.0713e-06 - acc: 1.0000 - val_loss: 0.0463 - val_acc: 0.9935
Epoch 20/250
48000/48000 [==============================] - 6s - loss: 8.8055e-07 - acc: 1.0000 - val_loss: 0.0465 - val_acc: 0.9936
Epoch 21/250
48000/48000 [==============================] - 6s - loss: 7.7420e-07 - acc: 1.0000 - val_loss: 0.0469 - val_acc: 0.9934
Epoch 22/250
48000/48000 [==============================] - 6s - loss: 6.4383e-07 - acc: 1.0000 - val_loss: 0.0472 - val_acc: 0.9934
Epoch 23/250
48000/48000 [==============================] - 6s - loss: 5.4621e-07 - acc: 1.0000 - val_loss: 0.0476 - val_acc: 0.9934
Epoch 24/250
48000/48000 [==============================] - 6s - loss: 4.7299e-07 - acc: 1.0000 - val_loss: 0.0479 - val_acc: 0.9934
Epoch 25/250
48000/48000 [==============================] - 6s - loss: 4.1923e-07 - acc: 1.0000 - val_loss: 0.0482 - val_acc: 0.9934
Epoch 26/250
48000/48000 [==============================] - 6s - loss: 3.6452e-07 - acc: 1.0000 - val_loss: 0.0485 - val_acc: 0.9934
Epoch 27/250
48000/48000 [==============================] - 6s - loss: 3.1688e-07 - acc: 1.0000 - val_loss: 0.0487 - val_acc: 0.9935
Epoch 28/250
48000/48000 [==============================] - 6s - loss: 2.8225e-07 - acc: 1.0000 - val_loss: 0.0490 - val_acc: 0.9934
Epoch 29/250
48000/48000 [==============================] - 6s - loss: 2.5075e-07 - acc: 1.0000 - val_loss: 0.0493 - val_acc: 0.9933
Epoch 30/250
48000/48000 [==============================] - 6s - loss: 2.2580e-07 - acc: 1.0000 - val_loss: 0.0496 - val_acc: 0.9933
Epoch 31/250
48000/48000 [==============================] - 6s - loss: 2.0984e-07 - acc: 1.0000 - val_loss: 0.0499 - val_acc: 0.9933
Epoch 32/250
48000/48000 [==============================] - 6s - loss: 1.9287e-07 - acc: 1.0000 - val_loss: 0.0501 - val_acc: 0.9933
Epoch 33/250
48000/48000 [==============================] - 6s - loss: 1.7815e-07 - acc: 1.0000 - val_loss: 0.0505 - val_acc: 0.9934
Epoch 34/250
48000/48000 [==============================] - 6s - loss: 1.6738e-07 - acc: 1.0000 - val_loss: 0.0507 - val_acc: 0.9933
Epoch 35/250
48000/48000 [==============================] - 6s - loss: 1.5871e-07 - acc: 1.0000 - val_loss: 0.0510 - val_acc: 0.9934
Epoch 36/250
48000/48000 [==============================] - 6s - loss: 1.5120e-07 - acc: 1.0000 - val_loss: 0.0514 - val_acc: 0.9933
Epoch 37/250
48000/48000 [==============================] - 6s - loss: 1.4508e-07 - acc: 1.0000 - val_loss: 0.0517 - val_acc: 0.9933
Epoch 38/250
48000/48000 [==============================] - 6s - loss: 1.3985e-07 - acc: 1.0000 - val_loss: 0.0518 - val_acc: 0.9934
Epoch 39/250
48000/48000 [==============================] - 6s - loss: 1.3628e-07 - acc: 1.0000 - val_loss: 0.0522 - val_acc: 0.9933
Epoch 40/250
48000/48000 [==============================] - 6s - loss: 1.3297e-07 - acc: 1.0000 - val_loss: 0.0525 - val_acc: 0.9933
Epoch 41/250
48000/48000 [==============================] - 6s - loss: 1.3011e-07 - acc: 1.0000 - val_loss: 0.0527 - val_acc: 0.9933
Epoch 42/250
48000/48000 [==============================] - 6s - loss: 1.2796e-07 - acc: 1.0000 - val_loss: 0.0531 - val_acc: 0.9933
Epoch 43/250
48000/48000 [==============================] - 6s - loss: 1.2638e-07 - acc: 1.0000 - val_loss: 0.0533 - val_acc: 0.9933
Epoch 44/250
48000/48000 [==============================] - 6s - loss: 1.2478e-07 - acc: 1.0000 - val_loss: 0.0536 - val_acc: 0.9933
Epoch 45/250
48000/48000 [==============================] - 6s - loss: 1.2356e-07 - acc: 1.0000 - val_loss: 0.0540 - val_acc: 0.9934
Epoch 46/250
48000/48000 [==============================] - 6s - loss: 1.2264e-07 - acc: 1.0000 - val_loss: 0.0543 - val_acc: 0.9934
Epoch 47/250
48000/48000 [==============================] - 6s - loss: 1.2195e-07 - acc: 1.0000 - val_loss: 0.0544 - val_acc: 0.9933
Epoch 48/250
48000/48000 [==============================] - 6s - loss: 1.2130e-07 - acc: 1.0000 - val_loss: 0.0548 - val_acc: 0.9933
Epoch 49/250
48000/48000 [==============================] - 6s - loss: 1.2081e-07 - acc: 1.0000 - val_loss: 0.0552 - val_acc: 0.9933
Epoch 50/250
48000/48000 [==============================] - 6s - loss: 1.2042e-07 - acc: 1.0000 - val_loss: 0.0554 - val_acc: 0.9933
Epoch 51/250
48000/48000 [==============================] - 6s - loss: 1.2011e-07 - acc: 1.0000 - val_loss: 0.0557 - val_acc: 0.9934
Epoch 52/250
48000/48000 [==============================] - 6s - loss: 1.1991e-07 - acc: 1.0000 - val_loss: 0.0559 - val_acc: 0.9933
Epoch 53/250
48000/48000 [==============================] - 6s - loss: 1.1971e-07 - acc: 1.0000 - val_loss: 0.0561 - val_acc: 0.9933
Epoch 54/250
48000/48000 [==============================] - 6s - loss: 1.1958e-07 - acc: 1.0000 - val_loss: 0.0562 - val_acc: 0.9933
Epoch 55/250
48000/48000 [==============================] - 6s - loss: 1.1946e-07 - acc: 1.0000 - val_loss: 0.0564 - val_acc: 0.9933
Epoch 56/250
48000/48000 [==============================] - 6s - loss: 1.1939e-07 - acc: 1.0000 - val_loss: 0.0567 - val_acc: 0.9933
Epoch 57/250
48000/48000 [==============================] - 6s - loss: 1.1933e-07 - acc: 1.0000 - val_loss: 0.0568 - val_acc: 0.9933
Epoch 58/250
48000/48000 [==============================] - 6s - loss: 1.1930e-07 - acc: 1.0000 - val_loss: 0.0571 - val_acc: 0.9933
Epoch 59/250
48000/48000 [==============================] - 6s - loss: 1.1927e-07 - acc: 1.0000 - val_loss: 0.0573 - val_acc: 0.9933
Epoch 60/250
48000/48000 [==============================] - 6s - loss: 1.1925e-07 - acc: 1.0000 - val_loss: 0.0574 - val_acc: 0.9933
Epoch 61/250
48000/48000 [==============================] - 6s - loss: 1.1924e-07 - acc: 1.0000 - val_loss: 0.0574 - val_acc: 0.9933
Epoch 62/250
48000/48000 [==============================] - 6s - loss: 1.1924e-07 - acc: 1.0000 - val_loss: 0.0577 - val_acc: 0.9934
Epoch 63/250
48000/48000 [==============================] - 6s - loss: 1.1922e-07 - acc: 1.0000 - val_loss: 0.0577 - val_acc: 0.9933
Epoch 64/250
48000/48000 [==============================] - 6s - loss: 1.1922e-07 - acc: 1.0000 - val_loss: 0.0579 - val_acc: 0.9933
Epoch 65/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0580 - val_acc: 0.9933
Epoch 66/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0580 - val_acc: 0.9933
Epoch 67/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0580 - val_acc: 0.9933
Epoch 68/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0580 - val_acc: 0.9933
Epoch 69/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0581 - val_acc: 0.9933
Epoch 70/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0582 - val_acc: 0.9933
Epoch 71/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0582 - val_acc: 0.9933
Epoch 72/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0582 - val_acc: 0.9933
Epoch 73/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0581 - val_acc: 0.9933
Epoch 74/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0582 - val_acc: 0.9933
Epoch 75/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0582 - val_acc: 0.9933
Epoch 76/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0582 - val_acc: 0.9933
Epoch 77/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 78/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 79/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 80/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 81/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 82/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 83/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 84/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 85/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 86/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 87/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 88/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 89/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 90/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 91/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 92/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 93/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 94/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 95/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 96/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 97/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 98/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 99/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 100/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 101/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 102/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 103/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 104/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 105/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 106/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 107/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 108/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 109/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 110/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 111/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 112/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 113/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 114/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 115/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 116/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 117/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 118/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 119/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 120/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 121/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 122/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 123/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 124/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 125/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 126/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 127/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 128/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 129/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 130/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 131/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 132/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 133/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 134/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 135/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 136/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 137/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 138/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 139/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 140/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 141/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 142/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 143/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 144/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 145/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 146/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 147/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 148/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 149/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 150/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 151/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 152/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 153/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 154/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 155/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 156/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 157/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 158/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 159/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 160/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 161/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 162/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 163/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 164/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 165/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 166/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 167/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 168/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 169/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 170/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 171/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 172/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 173/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 174/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 175/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 176/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 177/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 178/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 179/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 180/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 181/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 182/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 183/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 184/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 185/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 186/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 187/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 188/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 189/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 190/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 191/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 192/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 193/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 194/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 195/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 196/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 197/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 198/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 199/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 200/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 201/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 202/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 203/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 204/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 205/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 206/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 207/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 208/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 209/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 210/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 211/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 212/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 213/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 214/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 215/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 216/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 217/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 218/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 219/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 220/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 221/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 222/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 223/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 224/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 225/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 226/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 227/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 228/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 229/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 230/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 231/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 232/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 233/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 234/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 235/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 236/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 237/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 238/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 239/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 240/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 241/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 242/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 243/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 244/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 245/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 246/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 247/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 248/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 249/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Epoch 250/250
48000/48000 [==============================] - 6s - loss: 1.1921e-07 - acc: 1.0000 - val_loss: 0.0584 - val_acc: 0.9933
Step 10: Evaluate the model on the test dataset (10,000 images)
score = model.evaluate(X_test, Y_test, verbose=VERBOSE)
print("\nTest score:", score[0])
print('Test accuracy:', score[1])
9824/10000 [============================>.] - ETA: 0s
Test score: 0.051760694336
Test accuracy: 0.993
Step 11: Plot the accuracy from history
print(history.history.keys())
plt.plot(history.history['acc'])
plt.plot(history.history['val_acc'])
plt.title('model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper left')
plt.show()
dict_keys(['val_loss', 'val_acc', 'loss', 'acc'])
Step 12: Plot the loss from history
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('model loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper left')
plt.show()
[Optional] Step 13: Save the model (serialized) to JSON
model_json = model.to_json()
with open("model.json", "w") as json_file:
json_file.write(model_json)
%ls
Volume in drive C is Windows
Volume Serial Number is 4E4C-DF71
Directory of C:\Users\ECE\workspace
21/09/2017 10:47 AM <DIR> .
21/09/2017 10:47 AM <DIR> ..
21/09/2017 10:03 AM <DIR> .ipynb_checkpoints
22/07/2017 07:42 AM <DIR> appium-project
21/09/2017 10:03 AM 136,338 DeepLearningHelloWorldPart4.ipynb
21/09/2017 10:45 AM 139,720 DeepLearningHelloWorldPart5.ipynb
13/09/2017 08:10 PM <DIR> Deep-Learning-with-Keras
13/09/2017 05:03 PM <DIR> five-video-classification-methods
14/09/2017 06:51 AM <DIR> keras
21/09/2017 10:47 AM 3,146 model.json
3 File(s) 279,204 bytes
7 Dir(s) 63,157,465,088 bytes free
[Optional] Step 14: Save the model weights
model.save_weights("model.h5")
%ls
Volume in drive C is Windows
Volume Serial Number is 4E4C-DF71
Directory of C:\Users\ECE\workspace
21/09/2017 10:47 AM <DIR> .
21/09/2017 10:47 AM <DIR> ..
21/09/2017 10:03 AM <DIR> .ipynb_checkpoints
22/07/2017 07:42 AM <DIR> appium-project
21/09/2017 10:03 AM 136,338 DeepLearningHelloWorldPart4.ipynb
21/09/2017 10:47 AM 124,271 DeepLearningHelloWorldPart5.ipynb
13/09/2017 08:10 PM <DIR> Deep-Learning-with-Keras
13/09/2017 05:03 PM <DIR> five-video-classification-methods
14/09/2017 06:51 AM <DIR> keras
21/09/2017 10:47 AM 5,049,496 model.h5
21/09/2017 10:47 AM 3,146 model.json
4 File(s) 5,313,251 bytes
7 Dir(s) 63,152,427,008 bytes free
[Optional] Step 15: Load the saved model
json_file = open('model.json', 'r')
loaded_model_json = json_file.read()
json_file.close()
loaded_model = model_from_json(loaded_model_json)
loaded_model.load_weights("model.h5")
[Optional] Step 16: Compile and evaluate loaded model
loaded_model.compile(loss='categorical_crossentropy',
optimizer=OPTIMIZER,
metrics=['accuracy'])
score = loaded_model.evaluate(X_test, Y_test, verbose=VERBOSE)
print("\nTest score:", score[0])
print('Test accuracy:', score[1])
10000/10000 [==============================] - 149s
Test score: 0.051760694336
Test accuracy: 0.993
- mkc
Written on September 21, 2017