import torch
model = torch.hub.load('pytorch/vision:v0.10.0', 'vgg11', pretrained=True)
# or any of these variants
# model = torch.hub.load('pytorch/vision:v0.10.0', 'vgg11_bn', pretrained=True)
# model = torch.hub.load('pytorch/vision:v0.10.0', 'vgg13', pretrained=True)
# model = torch.hub.load('pytorch/vision:v0.10.0', 'vgg13_bn', pretrained=True)
# model = torch.hub.load('pytorch/vision:v0.10.0', 'vgg16', pretrained=True)
# model = torch.hub.load('pytorch/vision:v0.10.0', 'vgg16_bn', pretrained=True)
# model = torch.hub.load('pytorch/vision:v0.10.0', 'vgg19', pretrained=True)
# model = torch.hub.load('pytorch/vision:v0.10.0', 'vgg19_bn', pretrained=True)
model.eval()

所有预训练模型都要求输入图像以相同方式归一化，即形状为 (3 x H x W) 的 3 通道 RGB 图像的 mini-batch，其中 H 和 W 预期至少为 224。图像必须加载到 [0, 1] 范围内，然后使用 mean = [0.485, 0.456, 0.406] 和 std = [0.229, 0.224, 0.225] 进行归一化。

这是一个示例执行过程。

# Download an example image from the pytorch website
import urllib
url, filename = ("https://github.com/pytorch/hub/raw/master/images/dog.jpg", "dog.jpg")
try: urllib.URLopener().retrieve(url, filename)
except: urllib.request.urlretrieve(url, filename)

# sample execution (requires torchvision)
from PIL import Image
from torchvision import transforms
input_image = Image.open(filename)
preprocess = transforms.Compose([
    transforms.Resize(256),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])
input_tensor = preprocess(input_image)
input_batch = input_tensor.unsqueeze(0) # create a mini-batch as expected by the model

# move the input and model to GPU for speed if available
if torch.cuda.is_available():
    input_batch = input_batch.to('cuda')
    model.to('cuda')

with torch.no_grad():
    output = model(input_batch)
# Tensor of shape 1000, with confidence scores over ImageNet's 1000 classes
print(output[0])
# The output has unnormalized scores. To get probabilities, you can run a softmax on it.
probabilities = torch.nn.functional.softmax(output[0], dim=0)
print(probabilities)

# Download ImageNet labels
!wget https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt

# Read the categories
with open("imagenet_classes.txt", "r") as f:
    categories = [s.strip() for s in f.readlines()]
# Show top categories per image
top5_prob, top5_catid = torch.topk(probabilities, 5)
for i in range(top5_prob.size(0)):
    print(categories[top5_catid[i]], top5_prob[i].item())

模型描述

此处提供了论文《用于大规模图像识别的非常深度的卷积网络》中提出的模型的实现，包括每种配置及其带批量归一化的版本。

例如，论文中介绍的配置 A 是 vgg11，配置 B 是 vgg13，配置 D 是 vgg16，配置 E 是 vgg19。它们的批量归一化版本以 _bn 为后缀。

使用预训练模型在 ImageNet 数据集上的 Top-1 错误率如下表所示。

模型结构 Top-1 错误率 Top-5 错误率
vgg11 30.98 11.37
vgg11_bn 26.70 8.58
vgg13 30.07 10.75
vgg13_bn 28.45 9.63
vgg16 28.41 9.62
vgg16_bn 26.63 8.50
vgg19 27.62 9.12
vgg19_bn 25.76 8.15

参考文献