def svm_baseline():
img_images, img_labels, tst_images, tst_labels = main.load_mnist()
training_images, training_labels = util.subSample(img_images, img_labels,
50000)
training_data = (training_images, training_labels)
test_data = (tst_images, tst_labels)
print(training_data[0].shape)
print(training_data[1].shape)
print(test_data[0].shape)
print(test_data[1].shape)
# train
clf = svm.SVC()
clf.fit(training_data[0], training_data[1])
# test
predictions = [int(a) for a in clf.predict(test_data[0])]
num_correct = sum(int(a == y) for a, y in zip(predictions, test_data[1]))
print("Baseline classifier using an SVM.")
print("%s of %s values correct." % (num_correct, len(test_data[1])))
from main import training
from main import test
from main import load_mnist
img_images, img_labels, tst_images, tst_labels = load_mnist()
print(img_images.shape)
clf = training(img_images, img_labels, 500)
test_size = 100
num_correct = test(clf, tst_images, tst_labels, test_size)
print ("Baseline classifier using an SVM.")
print ("%s of %s values correct." % (num_correct, test_size))
parser.add_argument('--max_params', default=0, type=int)
parser.add_argument('chromosome', help="Chromosme")
args = parser.parse_args()
set_args(args)
print(args)
if not os.path.exists(args.save_dir):
os.makedirs(args.save_dir)
if not os.path.exists(args.cache_dir):
os.makedirs(args.cache_dir)
# load data
if args.dataset == 'mnist':
(x_train, y_train), (x_test, y_test) = load_mnist()
inshape = [28, 1, 1, 10]
elif args.dataset == 'fmnist' or args.dataset == 'fashion_mnist':
(x_train, y_train), (x_test, y_test) = load_fmnist()
inshape = [28, 1, 1, 10]
elif args.dataset == 'cifar10':
from keras.datasets import cifar10
(x_train, y_train), (x_test, y_test) = load_cifar10()
inshape = [32, 3, 1, 10]
elif args.dataset == 'cifar100':
from keras.datasets import cifar100
(x_train, y_train), (x_test, y_test) = load_cifar100()
inshape = [32, 3, 1, 100]
elif args.dataset == 'svhn':
(x_train, y_train), (x_test, y_test) = load_svhn()
inshape = [32, 3, 1, 10]
# -*- coding: utf-8 -*-
from main import load_mnist
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from KernelizedPerceptron import MyKP
if __name__ == '__main__':
images, labels = load_mnist('data/fashion')
x_test, y_test = load_mnist('data/fashion', kind='t10k')
x_train, x_validation, y_train, y_validation =\
train_test_split(images, labels, test_size=0.2)
degree = [2, 3, 4]
for i in degree:
MyKP(x_train, y_train, x_validation,
y_validation, x_test, y_test, i)