How to use the kymatio.datasets function in kymatio

To help you get started, we’ve selected a few kymatio examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

kymatio / kymatio / examples / 2d / cifar.py View on Github

pin_memory = False

    normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                     std=[0.229, 0.224, 0.225])

    train_loader = torch.utils.data.DataLoader(
        datasets.CIFAR10(root=scattering_datasets.get_dataset_dir('CIFAR'), train=True, transform=transforms.Compose([
            transforms.RandomHorizontalFlip(),
            transforms.RandomCrop(32, 4),
            transforms.ToTensor(),
            normalize,
        ]), download=True),
        batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)

    test_loader = torch.utils.data.DataLoader(
        datasets.CIFAR10(root=scattering_datasets.get_dataset_dir('CIFAR'), train=False, transform=transforms.Compose([
            transforms.ToTensor(),
            normalize,
        ])),
        batch_size=128, shuffle=False, num_workers=num_workers, pin_memory=pin_memory)

    # Optimizer
    lr = 0.1
    for epoch in range(0, 90):
        if epoch%20==0:
            optimizer = torch.optim.SGD(model.parameters(), lr=lr, momentum=0.9,
                                        weight_decay=0.0005)
            lr*=0.2

        train(model, device, train_loader, optimizer, epoch+1, scattering)
        test(model, device, test_loader, scattering)

kymatio / kymatio / examples / 2d / cifar_resnet.py View on Github

pin_memory = False

    normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                     std=[0.229, 0.224, 0.225])

    train_loader = torch.utils.data.DataLoader(
        datasets.CIFAR10(root=scattering_datasets.get_dataset_dir('CIFAR'), train=True, transform=transforms.Compose([
            transforms.RandomHorizontalFlip(),
            transforms.RandomCrop(32, 4),
            transforms.ToTensor(),
            normalize,
        ]), download=True),
        batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)

    test_loader = torch.utils.data.DataLoader(
        datasets.CIFAR10(root=scattering_datasets.get_dataset_dir('CIFAR'), train=False, transform=transforms.Compose([
            transforms.ToTensor(),
            normalize,
        ])),
        batch_size=128, shuffle=False, num_workers=num_workers, pin_memory=pin_memory)

    # Optimizer
    lr = 0.1
    for epoch in range(0, 90):
        if epoch%20==0:
            optimizer = torch.optim.SGD(model.parameters(), lr=lr, momentum=0.9,
                                        weight_decay=0.0005)
            lr*=0.2

        train(model, device, train_loader, optimizer, epoch+1, scattering)
        test(model, device, test_loader, scattering)

kymatio / kymatio / examples / 2d / mnist.py View on Github

if use_cuda:
        num_workers = 4
        pin_memory = True
    else:
        num_workers = None
        pin_memory = False

    train_loader = torch.utils.data.DataLoader(
        datasets.MNIST(scattering_datasets.get_dataset_dir('MNIST'), train=True, download=True,
                       transform=transforms.Compose([
                           transforms.ToTensor(),
                           transforms.Normalize((0.1307,), (0.3081,))
                       ])),
        batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)
    test_loader = torch.utils.data.DataLoader(
        datasets.MNIST(scattering_datasets.get_dataset_dir('MNIST'), train=False, transform=transforms.Compose([
            transforms.ToTensor(),
            transforms.Normalize((0.1307,), (0.3081,))
        ])),
        batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)

    # Optimizer
    optimizer = torch.optim.SGD(model.parameters(), lr=0.01, momentum=0.9,
                                weight_decay=0.0005)

    for epoch in range(1, 16):
        train( model, device, train_loader, optimizer, epoch, scattering)
        test(model, device, test_loader, scattering)

kymatio / kymatio / examples / 2d / cifar_resnet.py View on Github

model = Scattering2dResNet(K, args.width).to(device)

    # DataLoaders
    if use_cuda:
        num_workers = 4
        pin_memory = True
    else:
        num_workers = None
        pin_memory = False

    normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                     std=[0.229, 0.224, 0.225])

    train_loader = torch.utils.data.DataLoader(
        datasets.CIFAR10(root=scattering_datasets.get_dataset_dir('CIFAR'), train=True, transform=transforms.Compose([
            transforms.RandomHorizontalFlip(),
            transforms.RandomCrop(32, 4),
            transforms.ToTensor(),
            normalize,
        ]), download=True),
        batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)

    test_loader = torch.utils.data.DataLoader(
        datasets.CIFAR10(root=scattering_datasets.get_dataset_dir('CIFAR'), train=False, transform=transforms.Compose([
            transforms.ToTensor(),
            normalize,
        ])),
        batch_size=128, shuffle=False, num_workers=num_workers, pin_memory=pin_memory)

    # Optimizer
    lr = 0.1

kymatio / kymatio / examples / 2d / cifar.py View on Github

model = Scattering2dCNN(K,args.classifier).to(device)

    # DataLoaders
    if use_cuda:
        num_workers = 4
        pin_memory = True
    else:
        num_workers = None
        pin_memory = False

    normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                     std=[0.229, 0.224, 0.225])

    train_loader = torch.utils.data.DataLoader(
        datasets.CIFAR10(root=scattering_datasets.get_dataset_dir('CIFAR'), train=True, transform=transforms.Compose([
            transforms.RandomHorizontalFlip(),
            transforms.RandomCrop(32, 4),
            transforms.ToTensor(),
            normalize,
        ]), download=True),
        batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)

    test_loader = torch.utils.data.DataLoader(
        datasets.CIFAR10(root=scattering_datasets.get_dataset_dir('CIFAR'), train=False, transform=transforms.Compose([
            transforms.ToTensor(),
            normalize,
        ])),
        batch_size=128, shuffle=False, num_workers=num_workers, pin_memory=pin_memory)

    # Optimizer
    lr = 0.1

kymatio / kymatio / examples / 2d / mnist.py View on Github

m.weight.data.normal_(0, 2./math.sqrt(n))
            m.bias.data.zero_()
        if isinstance(m, nn.Linear):
            m.weight.data.normal_(0, 2./math.sqrt(m.in_features))
            m.bias.data.zero_()

    # DataLoaders
    if use_cuda:
        num_workers = 4
        pin_memory = True
    else:
        num_workers = None
        pin_memory = False

    train_loader = torch.utils.data.DataLoader(
        datasets.MNIST(scattering_datasets.get_dataset_dir('MNIST'), train=True, download=True,
                       transform=transforms.Compose([
                           transforms.ToTensor(),
                           transforms.Normalize((0.1307,), (0.3081,))
                       ])),
        batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)
    test_loader = torch.utils.data.DataLoader(
        datasets.MNIST(scattering_datasets.get_dataset_dir('MNIST'), train=False, transform=transforms.Compose([
            transforms.ToTensor(),
            transforms.Normalize((0.1307,), (0.3081,))
        ])),
        batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)

    # Optimizer
    optimizer = torch.optim.SGD(model.parameters(), lr=0.01, momentum=0.9,
                                weight_decay=0.0005)

kymatio / kymatio / examples / 2d / plot_mnist_classify_torch.py View on Github

prng = RandomState(42)

###############################################
# Create dataloaders
from torchvision import datasets, transforms
import kymatio.datasets as scattering_datasets

if use_cuda:
    num_workers = 4
    pin_memory = True
else:
    num_workers = 0
    pin_memory = False

train_data = datasets.MNIST(
                scattering_datasets.get_dataset_dir('MNIST'),
                train=True, download=True,
                transform=transforms.Compose([
                       transforms.ToTensor(),
                       transforms.Normalize((0.1307,), (0.3081,))
                ]))

#Extract a subset of 5000 samples from MNIST training
random_permute=prng.permutation(np.arange(0,60000))[0:5000]
train_data.data = train_data.data[random_permute]
train_data.targets = train_data.targets[random_permute]
train_loader = torch.utils.data.DataLoader(train_data,
    batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)

#Creat the test loader on the full MNIST test set
test_loader = torch.utils.data.DataLoader(
    datasets.MNIST(

kymatio / kymatio / examples / 2d / plot_mnist_classify_torch.py View on Github

transform=transforms.Compose([
                       transforms.ToTensor(),
                       transforms.Normalize((0.1307,), (0.3081,))
                ]))

#Extract a subset of 5000 samples from MNIST training
random_permute=prng.permutation(np.arange(0,60000))[0:5000]
train_data.data = train_data.data[random_permute]
train_data.targets = train_data.targets[random_permute]
train_loader = torch.utils.data.DataLoader(train_data,
    batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)

#Creat the test loader on the full MNIST test set
test_loader = torch.utils.data.DataLoader(
    datasets.MNIST(
        scattering_datasets.get_dataset_dir('MNIST'),
        train=False, transform=transforms.Compose([
        transforms.ToTensor(),
        transforms.Normalize((0.1307,), (0.3081,))
    ])),
    batch_size=128, shuffle=True, num_workers=num_workers, pin_memory=pin_memory)


###############################################################################
# This will help us define networks a bit more cleanly
import torch.nn as nn
class View(nn.Module):
    def __init__(self, *args):
        super(View, self).__init__()
        self.shape = args

    def forward(self, x):

How to use the kymatio.datasets function in kymatio

To help you get started, we’ve selected a few kymatio examples, based on popular ways it is used in public projects.

kymatio

Package Health Score

Popular kymatio functions

Similar packages