Pytorch官方教程(四)—Transfer Learning Tutorial
发布时间
阅读量：
阅读量
2019 April 04 pytorch
Pytorch官方教程(四)—Transfer Learning Tutorial
迁移学习主要的两种应用场景：
- 经过预训练的模型：我们采用经过预训练的模型来进行网络初始化，默认情况下不会进行随机初始化操作。具体而言，在基于Imagenet 1000的数据集上进行特定阶段的学习后（即经过特定优化后），剩余的部分则通过常规方式进行后续训练。
- 将卷积网络用作固定不变的特征提取器：通过固定权值不变的方式冻结除最后一层全连接层之外的所有网络参数权重。随后将最后一层全连接层替换为一个具有随机初始权重的新层，并对该新层进行参数优化。
# License: BSD
# Author: Sasank Chilamkurthy
from __future__ import print_function, division
import torch
import torch.nn as nn
import torch.optim as optim
from torch.optim import lr_scheduler
import numpy as np
import torchvision
from torchvision import datasets, models, transforms
import matplotlib.pyplot as plt
import time
import os
import copy
plt.ion() # interactive modeLoad Data
使用 torchvision 和 torch.utils.data 加载数据。
构建一种用于分类蚂蚁与蜜蜂的模型，在训练阶段中为每一类提供了120张图片；对应的验证集则包含每类75张图片。通常情况下, 若从零开始进行训练, 这样的数据量对于模型来说是极为有限的.通过采用迁移学习的方法, 可以显著提升模型的泛化性能.
这个数据集是imagenet的一个非常小的子集。
# Data augmentation and normalization for training
# Just normalization for validation
data_transforms = {
'train': transforms.Compose([
transforms.RandomResizedCrop(224),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
]),
'val': transforms.Compose([
transforms.Resize(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
]),
}
data_dir = 'data/hymenoptera_data'
image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x),
data_transforms[x])
for x in ['train', 'val']}
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=4,
shuffle=True, num_workers=4)
for x in ['train', 'val']}
dataset_sizes = {x: len(image_datasets[x]) for x in ['train', 'val']}
class_names = image_datasets['train'].classes
device = torch.device("cuda:2" if torch.cuda.is_available() else "cpu")Visualize a few images
可视化部分训练集的图片以便理解数据增强。
def imshow(inp, title=None):
"""Imshow for Tensor."""
inp = inp.numpy().transpose((1, 2, 0))
mean = np.array([0.485, 0.456, 0.406])
std = np.array([0.229, 0.224, 0.225])
inp = std * inp + mean
inp = np.clip(inp, 0, 1)
plt.imshow(inp)
if title is not None:
plt.title(title)
plt.pause(0.001) # pause a bit so that plots are updated
# Get a batch of training data
inputs, classes = next(iter(dataloaders['train']))
# Make a grid from batch
out = torchvision.utils.make_grid(inputs)
imshow(out, title=[class_names[x] for x in classes])
Training the model
定义模型训练的通用函数。需要声明：
- 学习率的调度
- 保存最好的模型
在代码中，默认使用的参数名为 scheduler 的源自 torch.optim.lr_scheduler 学习率调度器。
def train_model(model, criterion, optimizer, scheduler, num_epochs=25):
since = time.time()
best_model_wts = copy.deepcopy(model.state_dict())
best_acc = 0.0
for epoch in range(num_epochs):
print('Epoch {}/{}'.format(epoch, num_epochs - 1))
print('-' * 10)
# Each epoch has a training and validation phase
for phase in ['train', 'val']:
if phase == 'train':
scheduler.step()
model.train() # Set model to training mode
else:
model.eval() # Set model to evaluate mode
running_loss = 0.0
running_corrects = 0
# Iterate over data.
for inputs, labels in dataloaders[phase]:
inputs = inputs.to(device)
labels = labels.to(device)
# zero the parameter gradients
optimizer.zero_grad()
# forward
# track history if only in train
with torch.set_grad_enabled(phase == 'train'):
outputs = model(inputs)
_, preds = torch.max(outputs, 1)
loss = criterion(outputs, labels)
# backward + optimize only if in training phase
if phase == 'train':
loss.backward()
optimizer.step()
# statistics
running_loss += loss.item() * inputs.size(0)
running_corrects += torch.sum(preds == labels.data)
epoch_loss = running_loss / dataset_sizes[phase]
epoch_acc = running_corrects.double() / dataset_sizes[phase]
print('{} Loss: {:.4f} Acc: {:.4f}'.format(
phase, epoch_loss, epoch_acc))
# deep copy the model
if phase == 'val' and epoch_acc > best_acc:
best_acc = epoch_acc
best_model_wts = copy.deepcopy(model.state_dict())
print()
time_elapsed = time.time() - since
print('Training complete in {:.0f}m {:.0f}s'.format(
time_elapsed // 60, time_elapsed % 60))
print('Best val Acc: {:4f}'.format(best_acc))
# load best model weights
model.load_state_dict(best_model_wts)
return modelVisualizing the model predictions
显示预测的泛化函数
def visualize_model(model, num_images=6):
was_training = model.training
model.eval()
images_so_far = 0
fig = plt.figure()
with torch.no_grad():
for i, (inputs, labels) in enumerate(dataloaders['val']):
inputs = inputs.to(device)
labels = labels.to(device)
outputs = model(inputs)
_, preds = torch.max(outputs, 1)
for j in range(inputs.size()[0]):
images_so_far += 1
ax = plt.subplot(num_images//2, 2, images_so_far)
ax.axis('off')
ax.set_title('predicted: {}'.format(class_names[preds[j]]))
imshow(inputs.cpu().data[j])
if images_so_far == num_images:
model.train(mode=was_training)
return
model.train(mode=was_training)Finetuning the convnet
加载预训练的模型并重置最后的全连接层。
model_ft = models.resnet18(pretrained=True)
num_ftrs = model_ft.fc.in_features
model_ft.fc = nn.Linear(num_ftrs, 2)
model_ft = model_ft.to(device)
criterion = nn.CrossEntropyLoss()
# Observe that all parameters are being optimized
optimizer_ft = optim.SGD(model_ft.parameters(), lr=0.001, momentum=0.9)
# Decay LR by a factor of 0.1 every 7 epochs
exp_lr_scheduler = lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)Train and evaluate
在 CPU 上需 15-25 分钟. 在 GPU 不需要 1 分钟。
model_ft = train_model(model_ft, criterion, optimizer_ft, exp_lr_scheduler,
num_epochs=25) Epoch 0/24
----------
train Loss: 0.6469 Acc: 0.6393
val Loss: 0.2514 Acc: 0.9020
...
Epoch 24/24
----------
train Loss: 0.3563 Acc: 0.8320
val Loss: 0.2286 Acc: 0.9020
Training complete in 1m 5s
Best val Acc: 0.954248ConvNet as fixed feature extractor
在除了最后一层之外的所有层上固定参数。将所有可训练的神经网络节点标记为不计算梯度的状态，在反向传播过程中这些固定的参数就不会参与梯度的计算。
model_conv = torchvision.models.resnet18(pretrained=True)
for param in model_conv.parameters():
param.requires_grad = False
# Parameters of newly constructed modules have requires_grad=True by default
num_ftrs = model_conv.fc.in_features
model_conv.fc = nn.Linear(num_ftrs, 2)
model_conv = model_conv.to(device)
criterion = nn.CrossEntropyLoss()
# Observe that only parameters of final layer are being optimized as
# opposed to before.
optimizer_conv = optim.SGD(model_conv.fc.parameters(), lr=0.001, momentum=0.9)
# Decay LR by a factor of 0.1 every 7 epochs
exp_lr_scheduler = lr_scheduler.StepLR(optimizer_conv, step_size=7, gamma=0.1)Train and evaluate
网络的大部分梯度不需要计算，只需要计算前向传播。
model_conv = train_model(model_conv, criterion, optimizer_conv,
exp_lr_scheduler, num_epochs=25) Epoch 0/24
----------
train Loss: 0.5829 Acc: 0.6967
val Loss: 0.3998 Acc: 0.8039
...
Epoch 24/24
----------
train Loss: 0.3436 Acc: 0.8320
val Loss: 0.2410 Acc: 0.9020
Training complete in 0m 53s
Best val Acc: 0.954248 visualize_model(model_conv)
plt.ioff()
plt.show()
全部评论 (0)
还没有任何评论哟~