.. _sec_kaggle_cifar10:
Phân loại hình ảnh (CIFAR-10) trên Kaggle
=========================================
Cho đến nay, chúng tôi đã sử dụng các API cấp cao của các framework deep
learning để trực tiếp lấy tập dữ liệu hình ảnh ở định dạng tensor. Tuy
nhiên, các tập dữ liệu hình ảnh tùy chỉnh thường có dạng tệp hình ảnh.
Trong phần này, chúng ta sẽ bắt đầu từ các tệp hình ảnh thô và sắp xếp,
đọc, sau đó chuyển đổi chúng thành định dạng tensor từng bước.
Chúng tôi đã thử nghiệm với tập dữ liệu CIFAR-10 trong
:numref:`sec_image_augmentation`, đây là một tập dữ liệu quan trọng
trong tầm nhìn máy tính. Trong phần này, chúng tôi sẽ áp dụng kiến thức
chúng tôi đã học được trong các phần trước để thực hành cuộc thi Kaggle
của phân loại hình ảnh CIFAR-10. Địa chỉ web của cuộc thi là
https://www.kaggle.com/c/cifar-10
:numref:`fig_kaggle_cifar10` hiển thị thông tin trên trang web của
cuộc thi. Để gửi kết quả, bạn cần đăng ký tài khoản Kaggle.
.. _fig_kaggle_cifar10:
.. figure:: ../img/kaggle-cifar10.png
:width: 600px
CIFAR-10 image classification competition webpage information. The
competition dataset can be obtained by clicking the "Data" tab.
.. raw:: html
.. raw:: html
.. code:: python
import collections
import math
import os
import shutil
import pandas as pd
from mxnet import gluon, init, npx
from mxnet.gluon import nn
from d2l import mxnet as d2l
npx.set_np()
.. raw:: html
.. raw:: html
.. code:: python
import collections
import math
import os
import shutil
import pandas as pd
import torch
import torchvision
from torch import nn
from d2l import torch as d2l
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
#@save
d2l.DATA_HUB['cifar10_tiny'] = (d2l.DATA_URL + 'kaggle_cifar10_tiny.zip',
'2068874e4b9a9f0fb07ebe0ad2b29754449ccacd')
# If you use the full dataset downloaded for the Kaggle competition, set
# `demo` to False
demo = True
if demo:
data_dir = d2l.download_extract('cifar10_tiny')
else:
data_dir = '../data/cifar-10/'
.. parsed-literal::
:class: output
Downloading ../data/kaggle_cifar10_tiny.zip from http://d2l-data.s3-accelerate.amazonaws.com/kaggle_cifar10_tiny.zip...
.. raw:: html
.. raw:: html
.. code:: python
#@save
d2l.DATA_HUB['cifar10_tiny'] = (d2l.DATA_URL + 'kaggle_cifar10_tiny.zip',
'2068874e4b9a9f0fb07ebe0ad2b29754449ccacd')
# If you use the full dataset downloaded for the Kaggle competition, set
# `demo` to False
demo = True
if demo:
data_dir = d2l.download_extract('cifar10_tiny')
else:
data_dir = '../data/cifar-10/'
.. parsed-literal::
:class: output
Downloading ../data/kaggle_cifar10_tiny.zip from http://d2l-data.s3-accelerate.amazonaws.com/kaggle_cifar10_tiny.zip...
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
#@save
def read_csv_labels(fname):
"""Read `fname` to return a filename to label dictionary."""
with open(fname, 'r') as f:
# Skip the file header line (column name)
lines = f.readlines()[1:]
tokens = [l.rstrip().split(',') for l in lines]
return dict(((name, label) for name, label in tokens))
labels = read_csv_labels(os.path.join(data_dir, 'trainLabels.csv'))
print('# training examples:', len(labels))
print('# classes:', len(set(labels.values())))
.. parsed-literal::
:class: output
# training examples: 1000
# classes: 10
.. raw:: html
.. raw:: html
.. code:: python
#@save
def read_csv_labels(fname):
"""Read `fname` to return a filename to label dictionary."""
with open(fname, 'r') as f:
# Skip the file header line (column name)
lines = f.readlines()[1:]
tokens = [l.rstrip().split(',') for l in lines]
return dict(((name, label) for name, label in tokens))
labels = read_csv_labels(os.path.join(data_dir, 'trainLabels.csv'))
print('# training examples:', len(labels))
print('# classes:', len(set(labels.values())))
.. parsed-literal::
:class: output
# training examples: 1000
# classes: 10
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
#@save
def copyfile(filename, target_dir):
"""Copy a file into a target directory."""
os.makedirs(target_dir, exist_ok=True)
shutil.copy(filename, target_dir)
#@save
def reorg_train_valid(data_dir, labels, valid_ratio):
"""Split the validation set out of the original training set."""
# The number of examples of the class that has the fewest examples in the
# training dataset
n = collections.Counter(labels.values()).most_common()[-1][1]
# The number of examples per class for the validation set
n_valid_per_label = max(1, math.floor(n * valid_ratio))
label_count = {}
for train_file in os.listdir(os.path.join(data_dir, 'train')):
label = labels[train_file.split('.')[0]]
fname = os.path.join(data_dir, 'train', train_file)
copyfile(fname, os.path.join(data_dir, 'train_valid_test',
'train_valid', label))
if label not in label_count or label_count[label] < n_valid_per_label:
copyfile(fname, os.path.join(data_dir, 'train_valid_test',
'valid', label))
label_count[label] = label_count.get(label, 0) + 1
else:
copyfile(fname, os.path.join(data_dir, 'train_valid_test',
'train', label))
return n_valid_per_label
.. raw:: html
.. raw:: html
.. code:: python
#@save
def copyfile(filename, target_dir):
"""Copy a file into a target directory."""
os.makedirs(target_dir, exist_ok=True)
shutil.copy(filename, target_dir)
#@save
def reorg_train_valid(data_dir, labels, valid_ratio):
"""Split the validation set out of the original training set."""
# The number of examples of the class that has the fewest examples in the
# training dataset
n = collections.Counter(labels.values()).most_common()[-1][1]
# The number of examples per class for the validation set
n_valid_per_label = max(1, math.floor(n * valid_ratio))
label_count = {}
for train_file in os.listdir(os.path.join(data_dir, 'train')):
label = labels[train_file.split('.')[0]]
fname = os.path.join(data_dir, 'train', train_file)
copyfile(fname, os.path.join(data_dir, 'train_valid_test',
'train_valid', label))
if label not in label_count or label_count[label] < n_valid_per_label:
copyfile(fname, os.path.join(data_dir, 'train_valid_test',
'valid', label))
label_count[label] = label_count.get(label, 0) + 1
else:
copyfile(fname, os.path.join(data_dir, 'train_valid_test',
'train', label))
return n_valid_per_label
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
#@save
def reorg_test(data_dir):
"""Organize the testing set for data loading during prediction."""
for test_file in os.listdir(os.path.join(data_dir, 'test')):
copyfile(os.path.join(data_dir, 'test', test_file),
os.path.join(data_dir, 'train_valid_test', 'test',
'unknown'))
.. raw:: html
.. raw:: html
.. code:: python
#@save
def reorg_test(data_dir):
"""Organize the testing set for data loading during prediction."""
for test_file in os.listdir(os.path.join(data_dir, 'test')):
copyfile(os.path.join(data_dir, 'test', test_file),
os.path.join(data_dir, 'train_valid_test', 'test',
'unknown'))
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
def reorg_cifar10_data(data_dir, valid_ratio):
labels = read_csv_labels(os.path.join(data_dir, 'trainLabels.csv'))
reorg_train_valid(data_dir, labels, valid_ratio)
reorg_test(data_dir)
.. raw:: html
.. raw:: html
.. code:: python
def reorg_cifar10_data(data_dir, valid_ratio):
labels = read_csv_labels(os.path.join(data_dir, 'trainLabels.csv'))
reorg_train_valid(data_dir, labels, valid_ratio)
reorg_test(data_dir)
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
batch_size = 32 if demo else 128
valid_ratio = 0.1
reorg_cifar10_data(data_dir, valid_ratio)
.. raw:: html
.. raw:: html
.. code:: python
batch_size = 32 if demo else 128
valid_ratio = 0.1
reorg_cifar10_data(data_dir, valid_ratio)
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
transform_train = gluon.data.vision.transforms.Compose([
# Scale the image up to a square of 40 pixels in both height and width
gluon.data.vision.transforms.Resize(40),
# Randomly crop a square image of 40 pixels in both height and width to
# produce a small square of 0.64 to 1 times the area of the original
# image, and then scale it to a square of 32 pixels in both height and
# width
gluon.data.vision.transforms.RandomResizedCrop(32, scale=(0.64, 1.0),
ratio=(1.0, 1.0)),
gluon.data.vision.transforms.RandomFlipLeftRight(),
gluon.data.vision.transforms.ToTensor(),
# Standardize each channel of the image
gluon.data.vision.transforms.Normalize([0.4914, 0.4822, 0.4465],
[0.2023, 0.1994, 0.2010])])
.. raw:: html
.. raw:: html
.. code:: python
transform_train = torchvision.transforms.Compose([
# Scale the image up to a square of 40 pixels in both height and width
torchvision.transforms.Resize(40),
# Randomly crop a square image of 40 pixels in both height and width to
# produce a small square of 0.64 to 1 times the area of the original
# image, and then scale it to a square of 32 pixels in both height and
# width
torchvision.transforms.RandomResizedCrop(32, scale=(0.64, 1.0),
ratio=(1.0, 1.0)),
torchvision.transforms.RandomHorizontalFlip(),
torchvision.transforms.ToTensor(),
# Standardize each channel of the image
torchvision.transforms.Normalize([0.4914, 0.4822, 0.4465],
[0.2023, 0.1994, 0.2010])])
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
transform_test = gluon.data.vision.transforms.Compose([
gluon.data.vision.transforms.ToTensor(),
gluon.data.vision.transforms.Normalize([0.4914, 0.4822, 0.4465],
[0.2023, 0.1994, 0.2010])])
.. raw:: html
.. raw:: html
.. code:: python
transform_test = torchvision.transforms.Compose([
torchvision.transforms.ToTensor(),
torchvision.transforms.Normalize([0.4914, 0.4822, 0.4465],
[0.2023, 0.1994, 0.2010])])
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
train_ds, valid_ds, train_valid_ds, test_ds = [
gluon.data.vision.ImageFolderDataset(
os.path.join(data_dir, 'train_valid_test', folder))
for folder in ['train', 'valid', 'train_valid', 'test']]
.. raw:: html
.. raw:: html
.. code:: python
train_ds, train_valid_ds = [torchvision.datasets.ImageFolder(
os.path.join(data_dir, 'train_valid_test', folder),
transform=transform_train) for folder in ['train', 'train_valid']]
valid_ds, test_ds = [torchvision.datasets.ImageFolder(
os.path.join(data_dir, 'train_valid_test', folder),
transform=transform_test) for folder in ['valid', 'test']]
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
train_iter, train_valid_iter = [gluon.data.DataLoader(
dataset.transform_first(transform_train), batch_size, shuffle=True,
last_batch='discard') for dataset in (train_ds, train_valid_ds)]
valid_iter = gluon.data.DataLoader(
valid_ds.transform_first(transform_test), batch_size, shuffle=False,
last_batch='discard')
test_iter = gluon.data.DataLoader(
test_ds.transform_first(transform_test), batch_size, shuffle=False,
last_batch='keep')
.. raw:: html
.. raw:: html
.. code:: python
train_iter, train_valid_iter = [torch.utils.data.DataLoader(
dataset, batch_size, shuffle=True, drop_last=True)
for dataset in (train_ds, train_valid_ds)]
valid_iter = torch.utils.data.DataLoader(valid_ds, batch_size, shuffle=False,
drop_last=True)
test_iter = torch.utils.data.DataLoader(test_ds, batch_size, shuffle=False,
drop_last=False)
.. raw:: html
.. raw:: html
.. raw:: html
Ở đây, chúng tôi xây dựng các khối còn lại dựa trên lớp ``HybridBlock``,
hơi khác so với việc triển khai được mô tả trong :numref:`sec_resnet`.
Điều này là để cải thiện hiệu quả tính toán.
.. code:: python
class Residual(nn.HybridBlock):
def __init__(self, num_channels, use_1x1conv=False, strides=1, **kwargs):
super(Residual, self).__init__(**kwargs)
self.conv1 = nn.Conv2D(num_channels, kernel_size=3, padding=1,
strides=strides)
self.conv2 = nn.Conv2D(num_channels, kernel_size=3, padding=1)
if use_1x1conv:
self.conv3 = nn.Conv2D(num_channels, kernel_size=1,
strides=strides)
else:
self.conv3 = None
self.bn1 = nn.BatchNorm()
self.bn2 = nn.BatchNorm()
def hybrid_forward(self, F, X):
Y = F.npx.relu(self.bn1(self.conv1(X)))
Y = self.bn2(self.conv2(Y))
if self.conv3:
X = self.conv3(X)
return F.npx.relu(Y + X)
Tiếp theo, chúng tôi xác định mô hình ResNet-18.
.. code:: python
def resnet18(num_classes):
net = nn.HybridSequential()
net.add(nn.Conv2D(64, kernel_size=3, strides=1, padding=1),
nn.BatchNorm(), nn.Activation('relu'))
def resnet_block(num_channels, num_residuals, first_block=False):
blk = nn.HybridSequential()
for i in range(num_residuals):
if i == 0 and not first_block:
blk.add(Residual(num_channels, use_1x1conv=True, strides=2))
else:
blk.add(Residual(num_channels))
return blk
net.add(resnet_block(64, 2, first_block=True),
resnet_block(128, 2),
resnet_block(256, 2),
resnet_block(512, 2))
net.add(nn.GlobalAvgPool2D(), nn.Dense(num_classes))
return net
Chúng tôi sử dụng khởi tạo Xavier được mô tả trong
:numref:`subsec_xavier` trước khi đào tạo bắt đầu.
.. code:: python
def get_net(devices):
num_classes = 10
net = resnet18(num_classes)
net.initialize(ctx=devices, init=init.Xavier())
return net
loss = gluon.loss.SoftmaxCrossEntropyLoss()
.. raw:: html
.. raw:: html
Chúng tôi xác định mô hình ResNet-18 được mô tả trong
:numref:`sec_resnet`.
.. code:: python
def get_net():
num_classes = 10
net = d2l.resnet18(num_classes, 3)
return net
loss = nn.CrossEntropyLoss(reduction="none")
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
def train(net, train_iter, valid_iter, num_epochs, lr, wd, devices, lr_period,
lr_decay):
trainer = gluon.Trainer(net.collect_params(), 'sgd',
{'learning_rate': lr, 'momentum': 0.9, 'wd': wd})
num_batches, timer = len(train_iter), d2l.Timer()
legend = ['train loss', 'train acc']
if valid_iter is not None:
legend.append('valid acc')
animator = d2l.Animator(xlabel='epoch', xlim=[1, num_epochs],
legend=legend)
for epoch in range(num_epochs):
metric = d2l.Accumulator(3)
if epoch > 0 and epoch % lr_period == 0:
trainer.set_learning_rate(trainer.learning_rate * lr_decay)
for i, (features, labels) in enumerate(train_iter):
timer.start()
l, acc = d2l.train_batch_ch13(
net, features, labels.astype('float32'), loss, trainer,
devices, d2l.split_batch)
metric.add(l, acc, labels.shape[0])
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[2],
None))
if valid_iter is not None:
valid_acc = d2l.evaluate_accuracy_gpus(net, valid_iter,
d2l.split_batch)
animator.add(epoch + 1, (None, None, valid_acc))
measures = (f'train loss {metric[0] / metric[2]:.3f}, '
f'train acc {metric[1] / metric[2]:.3f}')
if valid_iter is not None:
measures += f', valid acc {valid_acc:.3f}'
print(measures + f'\n{metric[2] * num_epochs / timer.sum():.1f}'
f' examples/sec on {str(devices)}')
.. raw:: html
.. raw:: html
.. code:: python
def train(net, train_iter, valid_iter, num_epochs, lr, wd, devices, lr_period,
lr_decay):
trainer = torch.optim.SGD(net.parameters(), lr=lr, momentum=0.9,
weight_decay=wd)
scheduler = torch.optim.lr_scheduler.StepLR(trainer, lr_period, lr_decay)
num_batches, timer = len(train_iter), d2l.Timer()
legend = ['train loss', 'train acc']
if valid_iter is not None:
legend.append('valid acc')
animator = d2l.Animator(xlabel='epoch', xlim=[1, num_epochs],
legend=legend)
net = nn.DataParallel(net, device_ids=devices).to(devices[0])
for epoch in range(num_epochs):
net.train()
metric = d2l.Accumulator(3)
for i, (features, labels) in enumerate(train_iter):
timer.start()
l, acc = d2l.train_batch_ch13(net, features, labels,
loss, trainer, devices)
metric.add(l, acc, labels.shape[0])
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[2],
None))
if valid_iter is not None:
valid_acc = d2l.evaluate_accuracy_gpu(net, valid_iter)
animator.add(epoch + 1, (None, None, valid_acc))
scheduler.step()
measures = (f'train loss {metric[0] / metric[2]:.3f}, '
f'train acc {metric[1] / metric[2]:.3f}')
if valid_iter is not None:
measures += f', valid acc {valid_acc:.3f}'
print(measures + f'\n{metric[2] * num_epochs / timer.sum():.1f}'
f' examples/sec on {str(devices)}')
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
devices, num_epochs, lr, wd = d2l.try_all_gpus(), 20, 0.02, 5e-4
lr_period, lr_decay, net = 4, 0.9, get_net(devices)
net.hybridize()
train(net, train_iter, valid_iter, num_epochs, lr, wd, devices, lr_period,
lr_decay)
.. parsed-literal::
:class: output
train loss 0.726, train acc 0.757, valid acc 0.438
370.7 examples/sec on [gpu(0), gpu(1)]
.. figure:: output_kaggle-cifar10_42a34e_126_1.svg
.. raw:: html
.. raw:: html
.. code:: python
devices, num_epochs, lr, wd = d2l.try_all_gpus(), 20, 2e-4, 5e-4
lr_period, lr_decay, net = 4, 0.9, get_net()
train(net, train_iter, valid_iter, num_epochs, lr, wd, devices, lr_period,
lr_decay)
.. parsed-literal::
:class: output
train loss 0.736, train acc 0.749, valid acc 0.422
898.8 examples/sec on [device(type='cuda', index=0), device(type='cuda', index=1)]
.. figure:: output_kaggle-cifar10_42a34e_129_1.svg
.. raw:: html
.. raw:: html
.. raw:: html
.. code:: python
net, preds = get_net(devices), []
net.hybridize()
train(net, train_valid_iter, None, num_epochs, lr, wd, devices, lr_period,
lr_decay)
for X, _ in test_iter:
y_hat = net(X.as_in_ctx(devices[0]))
preds.extend(y_hat.argmax(axis=1).astype(int).asnumpy())
sorted_ids = list(range(1, len(test_ds) + 1))
sorted_ids.sort(key=lambda x: str(x))
df = pd.DataFrame({'id': sorted_ids, 'label': preds})
df['label'] = df['label'].apply(lambda x: train_valid_ds.synsets[x])
df.to_csv('submission.csv', index=False)
.. parsed-literal::
:class: output
train loss 0.950, train acc 0.657
830.9 examples/sec on [gpu(0), gpu(1)]
.. figure:: output_kaggle-cifar10_42a34e_135_1.svg
.. raw:: html
.. raw:: html
.. code:: python
net, preds = get_net(), []
train(net, train_valid_iter, None, num_epochs, lr, wd, devices, lr_period,
lr_decay)
for X, _ in test_iter:
y_hat = net(X.to(devices[0]))
preds.extend(y_hat.argmax(dim=1).type(torch.int32).cpu().numpy())
sorted_ids = list(range(1, len(test_ds) + 1))
sorted_ids.sort(key=lambda x: str(x))
df = pd.DataFrame({'id': sorted_ids, 'label': preds})
df['label'] = df['label'].apply(lambda x: train_valid_ds.classes[x])
df.to_csv('submission.csv', index=False)
.. parsed-literal::
:class: output
train loss 0.755, train acc 0.721
1165.8 examples/sec on [device(type='cuda', index=0), device(type='cuda', index=1)]
.. figure:: output_kaggle-cifar10_42a34e_138_1.svg
.. raw:: html
.. raw:: html
.. raw:: html
- Chúng ta có thể sử dụng các mạng thần kinh phức tạp, nâng hình ảnh và
lập trình lai trong một cuộc thi phân loại hình ảnh.
.. raw:: html
.. raw:: html
- Chúng ta có thể sử dụng mạng thần kinh phức tạp và nâng hình ảnh
trong một cuộc thi phân loại hình ảnh.
.. raw:: html
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html
`Discussions `__
.. raw:: html
.. raw:: html