实际应用时可能比较想获取VGG中间层的输出,

那么就可以如下操作:

import numpy as np
import torch
from torchvision import models
from torch.autograd import Variable
import torchvision.transforms as transforms
 
 
class CNNShow():
  def __init__(self, model):
    self.model = model
    self.model.eval()
 
    self.created_image = self.image_for_pytorch(np.uint8(np.random.uniform(150, 180, (224, 224, 3))))
 
 
  def show(self):
    x = self.created_image
    for index, layer in enumerate(self.model):
      print(index,layer)
      x = layer(x)
 
  def image_for_pytorch(self,Data):
    transform = transforms.Compose([
      transforms.ToTensor(), # range [0, 255] -> [0.0,1.0]
      transforms.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225))
    ]
    )
    imData = transform(Data)
    imData = Variable(torch.unsqueeze(imData, dim=0), requires_grad=True)
    return imData
 
if __name__ == '__main__':
 
  pretrained_model = models.vgg16(pretrained=True).features
  CNN = CNNShow(pretrained_model)
  CNN.show()

以上这篇pytorch获取vgg16-feature层输出的例子就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持。

广告合作:本站广告合作请联系QQ:858582 申请时备注:广告合作(否则不回)
免责声明:本站资源来自互联网收集,仅供用于学习和交流,请遵循相关法律法规,本站一切资源不代表本站立场,如有侵权、后门、不妥请联系本站删除!