from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVR
from helper import get_data, split_data, visualize
name = 'Support vector'
if __name__ == '__main__':
x, y = get_data()
y = y.reshape(len(y), 1)
x_train, x_test, y_train, y_test = split_data(x, y)
x_scaler = StandardScaler()
y_scaler = StandardScaler()
x_train = x_scaler.fit_transform(x_train)
x_test = x_scaler.transform(x_test)
y_train = y_scaler.fit_transform(y_train)
y_test = y_scaler.transform(y_test)
regression = SVR(kernel='rbf')
regression.fit(x_train, y_train)
y_predicted = regression.predict(x_test)
y_predicted = y_scaler.inverse_transform(y_predicted)
y_test = y_scaler.inverse_transform(y_test)
visualize(y_test, y_predicted, name)
Linear_test_aug = transforms.Compose(
[transforms.ToPILImage(),
transforms.ToTensor()])
Linear_testset = ImageDataset(root_dir=test_root,
class_file=classFile,
transforms=Linear_test_aug)
Linear_testloader = DataLoader(Linear_testset,
batch_size=512,
shuffle=False,
num_workers=num_worker)
# ========== [visualize] ==========
if batch_size >= 64:
visualize(Linear_trainloader, dir_log + '/' + 'visual.png')
# ========== [device] =============
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# ========== [cnn model] ==========
ckpt = torch.load(dir_ckpt + '/' + 'best.pt')
model = ResNet(pretrain=False)
model.load_state_dict(ckpt["cnn"])
model.to(device)
linear_clf = Linear_Classifier(classNum=10)
linear_clf.load_state_dict(ckpt["clf"])
linear_clf.to(device)
# opt_clf = optim.SGD(linear_clf.parameters(),
# lr=1e-2,
# momentum=0.9,
f_ishift = np.fft.ifftshift(s)
img_back = np.fft.ifft2(f_ishift)
img_back = np.abs(img_back)
fig, ax = plt.subplots(figsize=(20, 10))
plt.subplot(141), plt.imshow(img.numpy(), cmap='gray')
plt.title('Input Image'), plt.xticks([]), plt.yticks([])
plt.subplot(142), plt.imshow(magnitude_spectrum, cmap='gray')
plt.title('unmasked k-space'), plt.xticks([]), plt.yticks([])
plt.subplot(143), plt.imshow(20 * np.log(np.abs(s)), cmap='gray')
plt.title('masked k-space'), plt.xticks([]), plt.yticks([])
plt.subplot(144), plt.imshow(img_back, cmap='gray')
plt.title('ifft'), plt.xticks([]), plt.yticks([])
plt.show()
helper.visualize(helper.difference(img.numpy(), img_back))
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print(device)
transform = transforms.Compose([
transforms.Resize(256),
transforms.Grayscale(num_output_channels=1),
transforms.ToTensor()
])
batch_size = 20
# load the training and test datasets
train_data = datasets.ImageFolder(
root=
'C:\\Users\\tobi9\\Documents\\GitHub\\RefineGAN\\data\\brain\\db_train',
transform=transform)
test_data = datasets.ImageFolder(
from sklearn import svm import numpy import pylab from helper import visualize X=numpy.array([[0,0],[1,1],[1,0],[0,1]]) Y=[0,1,1,1] svc=svm.SVC(C=1.0,kernel='linear') trainedsvm=svc.fit(X,Y) visualize(trainedsvm,X,Y,[[-1.5,1.5],[-1.5,1.5]],100) print trainedsvm.predict([0,0]) print trainedsvm.predict([1,0])
Linear_testset = ImageDataset(
root_dir=test_root,
class_file=classFile,
transforms=Linear_test_aug
)
Linear_testloader = DataLoader(
Linear_testset,
batch_size=256,
shuffle=False,
num_workers=num_worker
)
# ========== [visualize] ==========
if batch_size >= 64:
visualize(trainloader, dir_log + '/' + 'visual.png')
# ========== [device] =============
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# ========== [cnn model] ==========
model = Resnet18()
model.to(device)
g = Projector(input_size=project_in, hidden_size=project_hidden, output_size=project_out)
g.to(device)
# ========== [optim for cnn] ==========
opt_cnn = optim.SGD(
list(model.parameters()) + list(g.parameters()),
lr=lr,
momentum=momentnum,
from sklearn import svm
import numpy
import pylab
from helper import visualize ,dataFromFile
obj1=dataFromFile('object1.txt')
obj2=dataFromFile('object2.txt')
label1= [1]*len(obj1)
label2=[0]*len(obj2)
trainingData=numpy.array( obj1+obj2)
labels=numpy.array(label1+label2)
svc=svm.SVC(C=1.0,kernel='linear')
trainedsvm=svc.fit(trainingData,labels)
visualize(trainedsvm,trainingData,labels,[[-1,25],[-1,25]],100)
#X=numpy.array([[0,0],[1,1],[1,0],[0,1]])
#Y=[0,1,1,1]
#trainedsvm=svc.fit(X,Y)
#print "True Label for [0,0] :" ,0, " Predicted :" ,trainedsvm.predict([0,0])
#print "True Label for [1,0:" ,1, " Predicted :" ,trainedsvm.predict([1,0])
#visualize(trainedsvm,X,Y,[[-1.5,1.5],[-1.5,1.5]],100)
#sample_weight=[2,1,1,1]
M = cv2.getRotationMatrix2D((cols / 2, rows / 2), 90, 1)
img_rot = np.rot90(img)
img_180 = np.rot90(img_rot)
img_rot = [[1, 1, 1, 0], [1, 1, 1, 0], [1, 1, 1, 0], [1, 1, 1, 0]]
img_180 = np.rot90(img_rot)
f_90 = np.fft.fft2(img_rot)
fshift_90 = np.fft.fftshift(f_90)
magnitude_spectrum_90 = (20 * np.log(np.abs(fshift_90)))
f_180 = np.fft.fft2(img_180)
fshift_180 = np.fft.fftshift(f_180)
magnitude_spectrum_180 = (20 * np.log(np.abs(fshift_180)))
helper.visualize(helper.difference(img, img_back))
#helper.visualize(helper.difference(magnitude_spectrum,magnitude_spectrum_180))
fig, ax = plt.subplots(figsize=(20, 10))
plt.subplot(141), plt.imshow(img, cmap='gray')
plt.title('Input Image'), plt.xticks([]), plt.yticks([])
plt.subplot(142), plt.imshow(magnitude_spectrum, cmap='gray')
plt.title('unmasked k-space'), plt.xticks([]), plt.yticks([])
plt.subplot(143), plt.imshow(ims, cmap='gray')
plt.title('masked k-space'), plt.xticks([]), plt.yticks([])
plt.subplot(144), plt.imshow(img_back, cmap='gray')
plt.title('ifft'), plt.xticks([]), plt.yticks([])
plt.show()
fig, ax = plt.subplots(figsize=(20, 10))
plt.subplot(141), plt.imshow(img_rot, cmap='gray')
return t, total
def info(title):
print(title)
print('module name:', __name__)
print('parent process:', os.getppid())
print('process id:', os.getpid())
if __name__ == '__main__':
info('main line')
# make larger file
file_iterations('filedata.txt', 100000, 'newfile.txt')
# # of processes
n = [1, 2, 3, 4, 5]
# times of processing, and total size of data streamed into buffer
times = []
sizes = []
for i in n:
result = test_mfmb(i)
times.append(result[0])
sizes.append(result[1])
visualize(times, n, sizes)
