def read_dataset(size_training, size_testing):
digits = [0,1,2,3,4,5,6,7,8,9]
images_train, labels_train = load_mnist('training',digits)
images_test, labels_test = load_mnist('testing',digits)
total_training = len(labels_train)
if(size_training > total_training):
size_training = total_training
total_testing = len(labels_test)
if(size_testing > total_testing):
size_testing = total_testing
random_training_instances = list(range(total_training))
random.shuffle(random_training_instances)
random_testing_instances = list(range(total_testing))
random.shuffle(random_training_instances)
images_train = images_train.astype(float64)
images_test = images_test.astype(float64)
images_train = images_train[random_training_instances[0:(size_training)],:]
labels_train = labels_train[random_training_instances[0:(size_training)]]
images_test = images_test[random_testing_instances[0:(size_testing)],:]
labels_test = labels_test[random_testing_instances[0:(size_testing)]]
return images_train, labels_train, images_test, labels_test
def read_dataset(size_training, size_testing):
digits = [0,1,2,3,4,5,6,7,8,9]
images_train, labels_train = load_mnist('training',digits)
images_test, labels_test = load_mnist('testing',digits)
total_training = len(labels_train)
if(size_training > total_training):
size_training = total_training
total_testing = len(labels_test)
if(size_testing > total_testing):
size_testing = total_testing
random_training_instances = range(total_training)
random.shuffle(random_training_instances)
random_testing_instances = range(total_testing)
random.shuffle(random_training_instances)
images_train = images_train.astype(float64)
images_test = images_test.astype(float64)
images_train = images_train[random_training_instances[0:(size_training)],:]
labels_train = labels_train[random_training_instances[0:(size_training)]]
images_test = images_test[random_testing_instances[0:(size_testing)],:]
labels_test = labels_test[random_testing_instances[0:(size_testing)]]
return images_train, labels_train, images_test, labels_test
def read_dataset(taille_apprentissage, taille_test,size_tot):
images, labels = load_mnist('training', digits=[0,1,2,3,4,5,6,7,8,9])
images_test, labels_test = load_mnist('testing', digits=[0,1,2,3,4,5,6,7,8,9])
images=images.astype(float64)
images_test=images_test.astype(float64)
images=images[1:(taille_apprentissage+1),:,:]
labels=labels[1:(taille_apprentissage+1)]
images_test=images_test[1:(taille_test+1),:,:]
labels_test=labels_test[1:(taille_test+1)]
return images, labels, images_test, labels_test
def read_dataset(taille_apprentissage, taille_test, size_tot):
images, labels = load_mnist('training',
digits=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
images_test, labels_test = load_mnist(
'testing', digits=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
images = images.astype(float64)
images_test = images_test.astype(float64)
images = images[1:(taille_apprentissage + 1), :, :]
labels = labels[1:(taille_apprentissage + 1)]
images_test = images_test[1:(taille_test + 1), :, :]
labels_test = labels_test[1:(taille_test + 1)]
return images, labels, images_test, labels_test
def read_dataset(taille_test,size_tot):
images_test, labels_test = load_mnist('testing', digits=[0,1,2,3,4,5,6,7,8,9])
images_test=images_test.astype(float64)
images_test=images_test[1:(taille_test+1),:,:]
labels_test=labels_test[1:(taille_test+1)]
return images_test, labels_test