.. _sec_image_augmentation:
画像拡張
========
| :numref:`sec_alexnet`
では、さまざまな応用において深層ニューラルネットワークが成功するための前提条件として、大規模データセットが必要であることを述べた。
| *画像拡張*
は、訓練画像に一連のランダムな変換を施すことで、元の画像に似ているが異なる訓練例を生成し、訓練データセットの規模を拡大する。
| 別の見方をすると、画像拡張は、訓練例に対するランダムな微調整によって、モデルが特定の属性に過度に依存しにくくなり、その結果として汎化能力が向上する、という事実に動機づけられている。
| たとえば、画像をさまざまな方法で切り抜くことで、注目対象が異なる位置に現れるようにでき、これにより対象の位置へのモデルの依存を減らせる。
| また、明るさや色などの要素を調整して、色に対するモデルの感度を下げることもできる。
| 当時の AlexNet の成功には、画像拡張が不可欠だった可能性が高いだろう。
| この節では、コンピュータビジョンで広く使われているこの手法について説明する。
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
%matplotlib inline
from d2l import torch as d2l
import torch
import torchvision
from torch import nn
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
%matplotlib inline
from d2l import mxnet as d2l
from mxnet import autograd, gluon, image, init, np, npx
from mxnet.gluon import nn
npx.set_np()
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
def apply(img, aug, num_rows=2, num_cols=4, scale=1.5):
Y = [aug(img) for _ in range(num_rows * num_cols)]
d2l.show_images(Y, num_rows, num_cols, scale=scale)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
def apply(img, aug, num_rows=2, num_cols=4, scale=1.5):
Y = [aug(img) for _ in range(num_rows * num_cols)]
d2l.show_images(Y, num_rows, num_cols, scale=scale)
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
d2l.set_figsize()
img = d2l.Image.open('../img/cat1.jpg')
d2l.plt.imshow(img);
.. figure:: output_image-augmentation_5ce40d_18_0.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
d2l.set_figsize()
img = image.imread('../img/cat1.jpg')
d2l.plt.imshow(img.asnumpy());
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
[07:00:45] ../src/storage/storage.cc:196: Using Pooled (Naive) StorageManager for CPU
.. figure:: output_image-augmentation_5ce40d_21_1.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
train_with_data_aug(train_augs, test_augs, net)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
train_with_data_aug(train_augs, test_augs, net)
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
apply(img, torchvision.transforms.RandomHorizontalFlip())
.. figure:: output_image-augmentation_5ce40d_35_0.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
apply(img, gluon.data.vision.transforms.RandomFlipLeftRight())
.. figure:: output_image-augmentation_5ce40d_38_0.svg
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
apply(img, torchvision.transforms.RandomVerticalFlip())
.. figure:: output_image-augmentation_5ce40d_44_0.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
apply(img, gluon.data.vision.transforms.RandomFlipTopBottom())
.. figure:: output_image-augmentation_5ce40d_47_0.svg
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
shape_aug = torchvision.transforms.RandomResizedCrop(
(200, 200), scale=(0.1, 1), ratio=(0.5, 2))
apply(img, shape_aug)
.. figure:: output_image-augmentation_5ce40d_53_0.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
shape_aug = gluon.data.vision.transforms.RandomResizedCrop(
(200, 200), scale=(0.1, 1), ratio=(0.5, 2))
apply(img, shape_aug)
.. figure:: output_image-augmentation_5ce40d_56_0.svg
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
apply(img, torchvision.transforms.ColorJitter(
brightness=0.5, contrast=0, saturation=0, hue=0))
.. figure:: output_image-augmentation_5ce40d_62_0.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
apply(img, gluon.data.vision.transforms.RandomBrightness(0.5))
.. figure:: output_image-augmentation_5ce40d_65_0.svg
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
apply(img, torchvision.transforms.ColorJitter(
brightness=0, contrast=0, saturation=0, hue=0.5))
.. figure:: output_image-augmentation_5ce40d_71_0.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
apply(img, gluon.data.vision.transforms.RandomHue(0.5))
.. figure:: output_image-augmentation_5ce40d_74_0.svg
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
color_aug = torchvision.transforms.ColorJitter(
brightness=0.5, contrast=0.5, saturation=0.5, hue=0.5)
apply(img, color_aug)
.. figure:: output_image-augmentation_5ce40d_80_0.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
color_aug = gluon.data.vision.transforms.RandomColorJitter(
brightness=0.5, contrast=0.5, saturation=0.5, hue=0.5)
apply(img, color_aug)
.. figure:: output_image-augmentation_5ce40d_83_0.svg
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
augs = torchvision.transforms.Compose([
torchvision.transforms.RandomHorizontalFlip(), color_aug, shape_aug])
apply(img, augs)
.. figure:: output_image-augmentation_5ce40d_89_0.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
augs = gluon.data.vision.transforms.Compose([
gluon.data.vision.transforms.RandomFlipLeftRight(), color_aug, shape_aug])
apply(img, augs)
.. figure:: output_image-augmentation_5ce40d_92_0.svg
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
all_images = torchvision.datasets.CIFAR10(train=True, root="../data",
download=True)
d2l.show_images([all_images[i][0] for i in range(32)], 4, 8, scale=0.8);
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ../data/cifar-10-python.tar.gz
100%|██████████| 170498071/170498071 [02:37<00:00, 1084703.38it/s]
Extracting ../data/cifar-10-python.tar.gz to ../data
.. figure:: output_image-augmentation_5ce40d_98_1.svg
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
d2l.show_images(gluon.data.vision.CIFAR10(
train=True)[:32][0], 4, 8, scale=0.8);
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
Downloading /opt/mxnet/datasets/cifar10/cifar-10-binary.tar.gz from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/dataset/cifar10/cifar-10-binary.tar.gz...
.. figure:: output_image-augmentation_5ce40d_101_1.svg
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
train_augs = torchvision.transforms.Compose([
torchvision.transforms.RandomHorizontalFlip(),
torchvision.transforms.ToTensor()])
test_augs = torchvision.transforms.Compose([
torchvision.transforms.ToTensor()])
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
train_augs = gluon.data.vision.transforms.Compose([
gluon.data.vision.transforms.RandomFlipLeftRight(),
gluon.data.vision.transforms.ToTensor()])
test_augs = gluon.data.vision.transforms.Compose([
gluon.data.vision.transforms.ToTensor()])
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
def load_cifar10(is_train, augs, batch_size):
dataset = torchvision.datasets.CIFAR10(root="../data", train=is_train,
transform=augs, download=True)
dataloader = torch.utils.data.DataLoader(dataset, batch_size=batch_size,
shuffle=is_train, num_workers=d2l.get_dataloader_workers())
return dataloader
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
def load_cifar10(is_train, augs, batch_size):
return gluon.data.DataLoader(
gluon.data.vision.CIFAR10(train=is_train).transform_first(augs),
batch_size=batch_size, shuffle=is_train,
num_workers=d2l.get_dataloader_workers())
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
#@save
def train_batch_ch13(net, X, y, loss, trainer, devices):
"""Train for a minibatch with multiple GPUs (defined in Chapter 13)."""
if isinstance(X, list):
# Required for BERT fine-tuning (to be covered later)
X = [x.to(devices[0]) for x in X]
else:
X = X.to(devices[0])
y = y.to(devices[0])
net.train()
trainer.zero_grad()
pred = net(X)
l = loss(pred, y)
l.sum().backward()
trainer.step()
train_loss_sum = l.sum()
train_acc_sum = d2l.accuracy(pred, y)
return train_loss_sum, train_acc_sum
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
#@save
def train_batch_ch13(net, features, labels, loss, trainer, devices,
split_f=d2l.split_batch):
"""Train for a minibatch with multiple GPUs (defined in Chapter 13)."""
X_shards, y_shards = split_f(features, labels, devices)
with autograd.record():
pred_shards = [net(X_shard) for X_shard in X_shards]
ls = [loss(pred_shard, y_shard) for pred_shard, y_shard
in zip(pred_shards, y_shards)]
for l in ls:
l.backward()
# The `True` flag allows parameters with stale gradients, which is useful
# later (e.g., in fine-tuning BERT)
trainer.step(labels.shape[0], ignore_stale_grad=True)
train_loss_sum = sum([float(l.sum()) for l in ls])
train_acc_sum = sum(d2l.accuracy(pred_shard, y_shard)
for pred_shard, y_shard in zip(pred_shards, y_shards))
return train_loss_sum, train_acc_sum
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
#@save
def train_ch13(net, train_iter, test_iter, loss, trainer, num_epochs,
devices=d2l.try_all_gpus()):
"""Train a model with multiple GPUs (defined in Chapter 13)."""
timer, num_batches = d2l.Timer(), len(train_iter)
animator = d2l.Animator(xlabel='epoch', xlim=[1, num_epochs], ylim=[0, 1],
legend=['train loss', 'train acc', 'test acc'])
net = nn.DataParallel(net, device_ids=devices).to(devices[0])
for epoch in range(num_epochs):
# Sum of training loss, sum of training accuracy, no. of examples,
# no. of predictions
metric = d2l.Accumulator(4)
for i, (features, labels) in enumerate(train_iter):
timer.start()
l, acc = train_batch_ch13(
net, features, labels, loss, trainer, devices)
metric.add(l, acc, labels.shape[0], labels.numel())
timer.stop()
if (i + 1) % (num_batches // 5) == 0 or i == num_batches - 1:
animator.add(epoch + (i + 1) / num_batches,
(metric[0] / metric[2], metric[1] / metric[3],
None))
test_acc = d2l.evaluate_accuracy_gpu(net, test_iter)
animator.add(epoch + 1, (None, None, test_acc))
print(f'loss {metric[0] / metric[2]:.3f}, train acc '
f'{metric[1] / metric[3]:.3f}, test acc {test_acc:.3f}')
print(f'{metric[2] * num_epochs / timer.sum():.1f} examples/sec on '
f'{str(devices)}')
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
#@save
def train_ch13(net, train_iter, test_iter, loss, trainer, num_epochs,
devices=d2l.try_all_gpus(), split_f=d2l.split_batch):
"""Train a model with multiple GPUs (defined in Chapter 13)."""
timer, num_batches = d2l.Timer(), len(train_iter)
animator = d2l.Animator(xlabel='epoch', xlim=[1, num_epochs], ylim=[0, 1],
legend=['train loss', 'train acc', 'test acc'])
for epoch in range(num_epochs):
# Sum of training loss, sum of training accuracy, no. of examples,
# no. of predictions
metric = d2l.Accumulator(4)
for i, (features, labels) in enumerate(train_iter):
timer.start()
l, acc = train_batch_ch13(
net, features, labels, loss, trainer, devices, split_f)
metric.add(l, acc, labels.shape[0], labels.size)
timer.stop()
if (i + 1) % (num_batches // 5) == 0 or i == num_batches - 1:
animator.add(epoch + (i + 1) / num_batches,
(metric[0] / metric[2], metric[1] / metric[3],
None))
test_acc = d2l.evaluate_accuracy_gpus(net, test_iter, split_f)
animator.add(epoch + 1, (None, None, test_acc))
print(f'loss {metric[0] / metric[2]:.3f}, train acc '
f'{metric[1] / metric[3]:.3f}, test acc {test_acc:.3f}')
print(f'{metric[2] * num_epochs / timer.sum():.1f} examples/sec on '
f'{str(devices)}')
.. raw:: html
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
batch_size, devices, net = 256, d2l.try_all_gpus(), d2l.resnet18(10, 3)
net.apply(d2l.init_cnn)
def train_with_data_aug(train_augs, test_augs, net, lr=0.001):
train_iter = load_cifar10(True, train_augs, batch_size)
test_iter = load_cifar10(False, test_augs, batch_size)
loss = nn.CrossEntropyLoss(reduction="none")
trainer = torch.optim.Adam(net.parameters(), lr=lr)
net(next(iter(train_iter))[0])
train_ch13(net, train_iter, test_iter, loss, trainer, 10, devices)
.. raw:: html
.. raw:: html
.. raw:: latex
\diilbookstyleinputcell
.. code:: python
batch_size, devices, net = 256, d2l.try_all_gpus(), d2l.resnet18(10)
net.initialize(init=init.Xavier(), ctx=devices)
def train_with_data_aug(train_augs, test_augs, net, lr=0.001):
train_iter = load_cifar10(True, train_augs, batch_size)
test_iter = load_cifar10(False, test_augs, batch_size)
loss = gluon.loss.SoftmaxCrossEntropyLoss()
trainer = gluon.Trainer(net.collect_params(), 'adam',
{'learning_rate': lr})
train_ch13(net, train_iter, test_iter, loss, trainer, 10, devices)
.. raw:: latex
\diilbookstyleoutputcell
.. parsed-literal::
:class: output
[07:00:57] ../src/storage/storage.cc:196: Using Pooled (Naive) StorageManager for GPU
[07:00:57] ../src/storage/storage.cc:196: Using Pooled (Naive) StorageManager for GPU
.. raw:: html
.. raw:: html