在上一篇,输入图像经过一个卷积核,得到一个输出。这是我们设置卷积层参数in_channel=1,out_channel=1时(默认)的结果。当我们设置in_channel=1,out_channel=2时,卷积核就回有两个,卷积后的输出也有两个
先简单编写一个神经网络,加载数据,显示定义的神经网络,如下:
import torch
import torchvision
from torch import nn
from torch.nn import Conv2d
from torch.utils.data import DataLoader
dataset = torchvision.datasets.CIFAR10("fl/data",train=False,transform=torchvision.transforms.ToTensor(),download=True)
#64个数据打包为一组
dataloader = DataLoader(dataset,batch_size=64)
class Mymodel(nn.Module):
def __init__(self):
super(Mymodel,self).__init__()
self.conv1 = Conv2d(in_channels=3,out_channels=6,kernel_size=3,stride=1,padding=0)
def forward(self,x):
x = self.conv1(x) #把x放到卷积层convx1,经过该卷积层之后会得到一个输出
return x
mymodel = Mymodel()
print(mymodel)
结果:
对图片进行相关处理,并用tensorboard打开
import torch
import torchvision
from tensorboardX import SummaryWriter
from torch import nn
from torch.nn import Conv2d
from torch.utils.data import DataLoader
dataset = torchvision.datasets.CIFAR10("fl/data",train=False,transform=torchvision.transforms.ToTensor(),download=True)
#64个数据打包为一组
dataloader = DataLoader(dataset,batch_size=64)
class Mymodel(nn.Module):
def __init__(self):
super(Mymodel,self).__init__()
self.conv1 = Conv2d(in_channels=3,out_channels=6,kernel_size=3,stride=1,padding=0)
def forward(self,x):
x = self.conv1(x) #把x放到卷积层convx1,经过该卷积层之后会得到一个输出
return x
mymodel = Mymodel()
print(mymodel)
writer = SummaryWriter("../logs")
#把每张图片放到神经网络中查看
step = 0
for data in dataloader:
#获取数据中的图片,已经经过转化成tensor类型了
imgs,targets = data
output = mymodel(imgs)
#imgs是输入图片
print(imgs.shape)
print(output.shape)
#输入大小 torch.Size([64, 3, 32, 32])
writer.add_images("input",imgs,step)
#经过卷积后的输出大小 torch.Size([64, 6, 30, 30]),彩图是3通道,6个通道的化tensorboard无法显示
#把上面的变成3通道-->[xxx,3,30,30],xxx不知道多少时,设置为-1
output = torch.reshape(output,(-1,3,30,30))
writer.add_images("output",output,step)
step = step + 1
在terminal下,运行tensorboard --logdir=logs,运行有误的参考上一篇如何使用tensorboard
点击链接打开图: