Vaibhav Khandelwal vaibhavcodes

## numpy.ipynb

      
              1 file
            
          
              0 forks
            
          
                0 comments
              
            
              0 stars
            
          
                vaibhavcodes
                / numpy.ipynb
            
            
              Created
              August 18, 2019 14:29
            
              
                Numpy.ipynb
              
          
      Loading

      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## LeNet-5_Architecture_import_library.py
import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Conv2D, AvgPool2D, Flatten, Dense
from keras.optimizers import Adam
from keras.utils import to_categorical
from keras.utils.vis_utils import plot_model

## LeNet-5_Architecture_dataset_splitting.py
(train_x, train_y), (test_x, test_y) = mnist.load_data()

## LeNet-5_Architecture_shape_of_train_and_test.py
print("The size of train_x is: {}".format(train_x.shape))
print("The size of train_y is: {}".format(train_y.shape))

print("The size of test_x is: {}".format(test_x.shape))
print("The size of test_y is: {}".format(test_y.shape))

## LeNet-5_Architecture_reshaping_image.py
train_x = train_x.reshape(train_x.shape[0], 28, 28, 1)
test_x = test_x.reshape(test_x.shape[0], 28, 28, 1)

## LeNet-5_Architecture_Normalization.py
train_x = train_x/255.0
test_x = test_x/255.0

## LeNet-5_Architecture_OneHotEncoding.py
train_y = to_categorical(train_y, num_classes=10)
test_y = to_categorical(test_y, num_classes=10)

## LeNet-5_Architecture_Model_Creation.py
# Instanciate an empty model
model = Sequential()

# Adding a Convolution Layer C1
# Input shape = N = (28 x 28)
# No. of filters  = 6
# Filter size = f = (5 x 5)
# Padding = P = 0
# Strides = S = 1
# Size of each feature map in C1 is (N-f+2P)/S +1 = 28-5+1 = 24

## LeNet-5_Architecture_Compilation_and_Fitting_of_model.py
model.compile(loss='categorical_crossentropy', optimizer=Adam(), metrics=['accuracy'])

model.fit(train_x, train_y, batch_size=128, epochs=20, verbose=1, validation_data=(test_x, test_y))

## LeNet-5_Architecture_loss_and_accuracy.py
score = model.evaluate(test_x, test_y)

print('Test Loss:', score[0])
print('Test accuracy:', score[1])
	import keras
	from keras.datasets import mnist
	from keras.models import Sequential
	from keras.layers import Conv2D, AvgPool2D, Flatten, Dense
	from keras.optimizers import Adam
	from keras.utils import to_categorical
	from keras.utils.vis_utils import plot_model
	print("The size of train_x is: {}".format(train_x.shape))
	print("The size of train_y is: {}".format(train_y.shape))

	print("The size of test_x is: {}".format(test_x.shape))
	print("The size of test_y is: {}".format(test_y.shape))
	train_x = train_x.reshape(train_x.shape[0], 28, 28, 1)
	test_x = test_x.reshape(test_x.shape[0], 28, 28, 1)
	train_y = to_categorical(train_y, num_classes=10)
	test_y = to_categorical(test_y, num_classes=10)
	# Instanciate an empty model
	model = Sequential()

	# Adding a Convolution Layer C1
	# Input shape = N = (28 x 28)
	# No. of filters = 6
	# Filter size = f = (5 x 5)
	# Padding = P = 0
	# Strides = S = 1
	# Size of each feature map in C1 is (N-f+2P)/S +1 = 28-5+1 = 24
	model.compile(loss='categorical_crossentropy', optimizer=Adam(), metrics=['accuracy'])

	model.fit(train_x, train_y, batch_size=128, epochs=20, verbose=1, validation_data=(test_x, test_y))
	score = model.evaluate(test_x, test_y)

	print('Test Loss:', score[0])
	print('Test accuracy:', score[1])