Day #327 - Keras basic experiments

Revising basics again. Example code of MNIST in Sequential vs Functional vs Multi-Inputs based approach. Already we have posted a few best practices/guidelines notes. These are few more handly experiments.

	#https://github.com/keras-team/keras/blob/master/examples/mnist_cnn.py
	from __future__ import print_function
	import keras
	from keras.datasets import mnist
	from keras.models import Sequential
	from keras.layers import Dense, Dropout, Flatten, Input, concatenate
	from keras.layers import Conv2D, MaxPooling2D
	from keras.models import Model
	from keras import backend as K
	batch_size = 128
	num_classes = 10
	epochs = 3
	# input image dimensions
	img_rows, img_cols = 28, 28
	def load_data():
	# the data, split between train and test sets
	(x_train, y_train), (x_test, y_test) = mnist.load_data()
	if K.image_data_format() == 'channels_first':
	x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)
	x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)
	input_shape = (1, img_rows, img_cols)
	else:
	x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)
	x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)
	input_shape = (img_rows, img_cols, 1)
	x_train = x_train.astype('float32')
	x_test = x_test.astype('float32')
	x_train /= 255
	x_test /= 255
	print('x_train shape:', x_train.shape)
	print(x_train.shape[0], 'train samples')
	print(x_test.shape[0], 'test samples')
	# convert class vectors to binary class matrices
	y_train = keras.utils.to_categorical(y_train, num_classes)
	y_test = keras.utils.to_categorical(y_test, num_classes)
	return input_shape, x_train, x_test, y_train, y_test
	input_shape, x_train, x_test, y_train, y_test = load_data()
	def model_definition_sequence_model(input_shape,num_classes):
	model = Sequential()
	model.add(Conv2D(32, kernel_size=(3, 3),
	activation='relu',
	input_shape=input_shape))
	model.add(Conv2D(64, (3, 3), activation='relu'))
	model.add(MaxPooling2D(pool_size=(2, 2)))
	model.add(Dropout(0.25))
	model.add(Flatten())
	model.add(Dense(128, activation='relu'))
	model.add(Dropout(0.5))
	model.add(Dense(num_classes, activation='softmax'))
	model.compile(loss=keras.losses.categorical_crossentropy,
	optimizer=keras.optimizers.Adadelta(),
	metrics=['accuracy'])
	return model
	#base code
	#https://github.com/PacktPublishing/Advanced-Deep-Learning-with-Keras/blob/master/chapter2-deep-networks/cnn-functional-2.1.1.py
	def model_definition_functional_model(input_shape,num_classes):
	kernel_size = 3
	filters = 64
	dropout = 0.3
	inputs = Input(shape=input_shape)
	y = Conv2D(filters=filters,
	kernel_size=kernel_size,
	activation='relu')(inputs)
	y = MaxPooling2D()(y)
	y = Conv2D(filters=filters,
	kernel_size=kernel_size,
	activation='relu')(y)
	y = MaxPooling2D()(y)
	y = Conv2D(filters=filters,
	kernel_size=kernel_size,
	activation='relu')(y)
	# image to vector before connecting to dense layer
	y = Flatten()(y)
	# dropout regularization
	y = Dropout(dropout)(y)
	outputs = Dense(num_classes, activation='softmax')(y)
	# build the model by supplying inputs/outputs
	model = Model(inputs=inputs, outputs=outputs)
	# network model in text
	model.summary()
	model.compile(loss=keras.losses.categorical_crossentropy,
	optimizer=keras.optimizers.Adadelta(),
	metrics=['accuracy'])
	return model
	#base code
	#https://github.com/PacktPublishing/Advanced-Deep-Learning-with-Keras/blob/master/chapter2-deep-networks/cnn-y-network-2.1.2.py
	def branch_definition_functional_model(input_shape,num_classes):
	kernel_size = 3
	filters = 64
	dropout = 0.3
	left_inputs = Input(shape=input_shape)
	y1 = Conv2D(filters=filters,
	kernel_size=kernel_size,
	activation='relu')(left_inputs)
	y1 = MaxPooling2D()(y1)
	right_inputs = Input(shape=input_shape)
	y2 = Conv2D(filters=filters,
	kernel_size=kernel_size,
	activation='relu')(right_inputs)
	y2 = MaxPooling2D()(y2)
	y = concatenate([y1,y2])
	y = Conv2D(filters=filters,
	kernel_size=kernel_size,
	activation='relu')(y)
	y = MaxPooling2D()(y)
	y = Conv2D(filters=filters,
	kernel_size=kernel_size,
	activation='relu')(y)
	# image to vector before connecting to dense layer
	y = Flatten()(y)
	# dropout regularization
	y = Dropout(dropout)(y)
	outputs = Dense(num_classes, activation='softmax')(y)
	# build the model by supplying inputs/outputs
	model = Model(inputs=[left_inputs,right_inputs], outputs=outputs)
	# network model in text
	model.summary()
	model.compile(loss=keras.losses.categorical_crossentropy,
	optimizer=keras.optimizers.Adadelta(),
	metrics=['accuracy'])
	return model
	def seq_approach(input_shape,num_classes):
	model = model_definition_sequence_model(input_shape,num_classes)
	model.fit(x_train, y_train,
	batch_size=batch_size,
	epochs=epochs,
	verbose=1,
	validation_data=(x_test, y_test))
	score = model.evaluate(x_test, y_test, verbose=0)
	print('seq_approach Test loss:', score[0])
	print('seq_approach Test accuracy:', score[1])
	def fun_approach(input_shape,num_classes):
	model = model_definition_functional_model(input_shape,num_classes)
	model.fit(x_train, y_train,
	batch_size=batch_size,
	epochs=epochs,
	verbose=1,
	validation_data=(x_test, y_test))
	score = model.evaluate(x_test, y_test, verbose=0)
	print('fun_approach Test loss:', score[0])
	print('fun_approach Test accuracy:', score[1])
	def branch_approach(input_shape,num_classes):
	model = branch_definition_functional_model(input_shape,num_classes)
	#Two feature vectors
	#In future generate two feature vectors vgg/ resnet and build classification models
	model.fit([x_train,x_train], y_train,
	batch_size=batch_size,
	epochs=epochs,
	verbose=1,
	validation_data=([x_test,x_test], y_test))
	score = model.evaluate([x_test,x_test], y_test, verbose=0)
	print('branch_approach Test loss:', score[0])
	print('branch_approach Test accuracy:', score[1])
	seq_approach(input_shape,num_classes)
	fun_approach(input_shape,num_classes)
	branch_approach(input_shape,num_classes)
	#seq_approach Test loss: 0.03464194674291939
	#seq_approach Test accuracy: 0.9878
	#fun_approach Test loss: 0.030942173414956777
	#fun_approach Test accuracy: 0.9895
	#branch_approach Test loss: 0.031166232500807384
	#branch_approach Test accuracy: 0.9899
	#https://keras.io/layers/merge/
	#Layer that adds a list of inputs.
	#import keras
	#input1 = keras.layers.Input(shape=(16,))
	#x1 = keras.layers.Dense(8, activation='relu')(input1)
	#input2 = keras.layers.Input(shape=(32,))
	#x2 = keras.layers.Dense(8, activation='relu')(input2)
	# equivalent to added = keras.layers.add([x1, x2])
	#added = keras.layers.Add()([x1, x2])
	#out = keras.layers.Dense(4)(added)
	#model = keras.models.Model(inputs=[input1, input2], outputs=out)

view raw keras_examples.py hosted with ❤ by GitHub

Happy Learning!!!