pytorch如何轻松获取特征图

通过torchvision提供的create_feature_extractor函数，我们可以根据层的名称轻松获取特征图。

文章所选用的测试图像如下：

输出结果 ：

以下为全部代码：

import torch
import torchvision
import torchvision.transforms as transforms
from PIL import Image
from torchvision.models.feature_extraction import create_feature_extractor
import einops
import matplotlib.pyplot as plt

def read_image_as_tensor(name: str):
    preprocess = transforms.Compose([
        transforms.ToTensor(),  # 将图像转换为张量
        transforms.Resize((224, 224), antialias=True)
    ])

    return preprocess(Image.open(name)).unsqueeze(0)


features = ['layer2.0.conv1', "layer3.0.conv1"]

model = torchvision.models.resnet18(
    weights=torchvision.models.ResNet18_Weights.DEFAULT)

feature_extractor = create_feature_extractor(model, return_nodes=features)
# 定义输入
x = read_image_as_tensor('n0153282900000010.jpg')

out = feature_extractor(x)
print(out['layer2.0.conv1'].shape)
print(out['layer3.0.conv1'].shape)
f1 = out['layer3.0.conv1'] # 特征图
# 调整以方便可视化
f1 = torch.reshape(f1, (16, 16, 14, 14))
f1 = einops.rearrange(f1, "H W h w -> (H h)(W w)")
plt.imshow(f1.detach().numpy())
plt.show()

 那么如何知道模型中不同层的名称呢？一种方法是直接print打印模型，另一种方法是使用get_graph_node_names函数。

import torchvision
import torch
from torchvision.models.feature_extraction import get_graph_node_names

model = torchvision.models.resnet18(
    weights=torchvision.models.ResNet18_Weights.DEFAULT)
# 方法1
print(model)
# 方法2
nodes, _ = get_graph_node_names(model)
print(nodes)

这两种结果的输出分别如下：

ResNet(
  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (relu): ReLU(inplace=True)
  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
  (layer1): Sequential(
    (0): BasicBlock(
      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
    (1): BasicBlock(
      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (layer2): Sequential(
    (0): BasicBlock(
      (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (downsample): Sequential(
        (0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): BasicBlock(
      (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (layer3): Sequential(
    (0): BasicBlock(
      (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (downsample): Sequential(
        (0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): BasicBlock(
      (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (layer4): Sequential(
    (0): BasicBlock(
      (conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (downsample): Sequential(
        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): BasicBlock(
      (conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))
  (fc): Linear(in_features=512, out_features=1000, bias=True)
)

['x', 'conv1', 'bn1', 'relu', 'maxpool', 'layer1.0.conv1', 'layer1.0.bn1', 'layer1.0.relu', 'layer1.0.conv2', 'layer1.0.bn2', 'layer1.0.add', 'layer1.0.relu_1', 'layer1.1.conv1', 'layer1.1.bn1', 'layer1.1.relu', 'layer1.1.conv2', 'layer1.1.bn2', 'layer1.1.add', 'layer1.1.relu_1', 'layer2.0.conv1', 'layer2.0.bn1', 'layer2.0.relu', 'layer2.0.conv2', 'layer2.0.bn2', 'layer2.0.downsample.0', 'layer2.0.downsample.1', 'layer2.0.add', 'layer2.0.relu_1', 'layer2.1.conv1', 'layer2.1.bn1', 'layer2.1.relu', 'layer2.1.conv2', 'layer2.1.bn2', 'layer2.1.add', 'layer2.1.relu_1', 'layer3.0.conv1', 'layer3.0.bn1', 'layer3.0.relu', 'layer3.0.conv2', 'layer3.0.bn2', 'layer3.0.downsample.0', 'layer3.0.downsample.1', 'layer3.0.add', 'layer3.0.relu_1', 'layer3.1.conv1', 'layer3.1.bn1', 'layer3.1.relu', 'layer3.1.conv2', 'layer3.1.bn2', 'layer3.1.add', 'layer3.1.relu_1', 'layer4.0.conv1', 'layer4.0.bn1', 'layer4.0.relu', 'layer4.0.conv2', 'layer4.0.bn2', 'layer4.0.downsample.0', 'layer4.0.downsample.1', 'layer4.0.add', 'layer4.0.relu_1', 'layer4.1.conv1', 'layer4.1.bn1', 'layer4.1.relu', 'layer4.1.conv2', 'layer4.1.bn2', 'layer4.1.add', 'layer4.1.relu_1', 'avgpool', 'flatten', 'fc']