pytorch_vision_vgg.md

layout	background-class	body-class	title	summary	category	image	author	tags	github-link	featured_image_1	featured_image_2
pytorch_hub_detail	pytorch-hub-background	pytorch-hub	VGG	Award winning models in ILSVRC challenge 2014.	researchers	pytorch-logo.png	Pytorch Team	CV image classification	https://github.com/pytorch/vision.git	vgg.png	no-image

Model Description

Here we have implementations for the models proposed in Very Deep Convolutional Networks for Large-Scale Image Recognition, for each configurations and their with bachnorm version.

For example, configuration A presented in the paper is vgg11, configuration B is vgg13, configuration D is vgg16 and configuration E is vgg19. Their batchnorm version are suffixed with _bn.

Their 1-crop error rates on imagenet dataset with pretrained models are list below.

Model structure	Top-1 error	Top-5 error
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

Notes on Inputs

All pre-trained models expect input images normalized in the same way, i.e. mini-batches of 3-channel RGB images of shape (3 x H x W), where H and W are expected to be at least 224. The images have to be loaded in to a range of [0, 1] and then normalized using mean = [0.485, 0.456, 0.406] and std = [0.229, 0.224, 0.225]. You can use the following transform to normalize:

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

Example:

import torch
model = torch.hub.load('pytorch/vision', 'vgg11', pretrained=True)
model = torch.hub.load('pytorch/vision', 'vgg11_bn', pretrained=True)
model = torch.hub.load('pytorch/vision', 'vgg13', pretrained=True)
model = torch.hub.load('pytorch/vision', 'vgg13_bn', pretrained=True)
model = torch.hub.load('pytorch/vision', 'vgg16', pretrained=True)
model = torch.hub.load('pytorch/vision', 'vgg16_bn', pretrained=True)
model = torch.hub.load('pytorch/vision', 'vgg19', pretrained=True)
model = torch.hub.load('pytorch/vision', 'vgg19_bn', pretrained=True)

Resources:

• Very Deep Convolutional Networks for Large-Scale Image Recognition.