z1139269312的博客

AVE相当于把图片压缩在解压的感觉。自编码器 x ——编码器——z——解码器——变分自编码器 对z有一个正态分布的约束import osimport torchimport torch.nn as nnimport torch.nn.functional as Fimport torchvisionfrom torchvision import transformsfrom torchvision.utils import save_imagesampl...

tensorboard 可视化神经网络import torchimport torch.nn as nnimport torch.nn.functional as Ffrom torch.utils.tensorboard import SummaryWriterclass Net(nn.Module): def __init__(self): super(Net, self).__init__() self.conv1 = nn.Conv2d(i.

utils.data包括Dataset和DataLoader。torch.utils.data.Dataset为抽象类。自定义数据集余姚继承这个类，并实现两个函数：__getitem__ , __len__ ,前者通过给定的索引获取数据和标签，后者提供数据集大小。__getitem__ 一次只能获取一个数据，所以用DataLoader来实现batchsize的读取。import numpy as npimport torchfrom torch.utils import data..

import torchimport torch.utils.data as Dataimport torch.nn.functional as Ffrom matplotlib import pyplot as pltlr = 0.01batchsize = 32epoch = 12x = torch.unsqueeze(torch.linspace(-1,1,1000),dim=1)# unsqueeze 把一维变二维y = x.pow(2) + 0.1 * torch.norma.

import numpy as npimport torchfrom matplotlib import pyplot as pltfrom torchvision.datasets import mnistimport torchvision.transforms as transformsfrom torch.utils.data import DataLoaderimport torch.nn.functional as Fimport torch.optim as optimfro.

表达式 y = 3 + 2 加上一些噪音数据得到另一个数据y构建一个机器学习模型，学习表达式y = + b ,利用数组x，y的数据为训练数据采用梯度下降法，通过多次迭代，学习到w，b的值# 损失函数 Loss =0.5(+b - 自动计算梯度import torchfrom matplotlib import pyplot as plttorch.manual_seed(100)dtype = torch.floatx = torch.unsq...

表达式 y = 3+ 2 加上一些噪音数据得到另一个数据y构建一个机器学习模型，学习表达式y =+ b ,利用数组x，y的数据为训练数据采用梯度下降法，通过多次迭代，学习到w，b的值# 损失函数 Loss =0.5(+b - 对损失函数求导 , 梯度下降法 w1 = w - b1 = b -import numpy as npfrom matplotlib im...

vgg16因为vgg16太大了（100多个G）所以没有下载pretrained = False 模型中所有的参数为默认参数pretrained = True 模型中所的参数为训练好的参数from torch import nnvgg16_false = torchvision.models.vgg16(pretrained=False)vgg16_true = torchvision.models.vgg16(pretrained=True)pr...

import torchimport torchvisionfrom torch import nnfrom torch.nn import Conv2d, MaxPool2d, Flatten, Linear, Sequentialfrom torch.utils.data import DataLoaderfrom torch.utils.tensorboard import SummaryWriterdataset = torchvision.datasets.CIFAR10("./d.

注意输入和输出的大小import torchfrom torch import nnfrom torch.nn import L1Lossinputs = torch.tensor([1,2,3],dtype=torch.float32)targets = torch.tensor([1,2,5],dtype=torch.float32)inputs = torch.reshape(inputs,(1,1,1,3))targets = torch.reshape(targets,(1,

输入 ——卷积——非线性——卷积——非线性——输出 forwarddef forward(self,x): x = F.relu(self.conv1(x)) # conv第一次卷积 1是一维 relu第一次非线性 return F.relu(self.conv2(x))nn.Module的使用import torchfrom...

import torchvisionfrom torch.utils.data import DataLoaderfrom torch.utils.tensorboard import SummaryWritertest_data = torchvision.datasets.CIFAR10("./dataset",train=False,transform=torchvision.transforms.ToTensor())#准备的一个测试集test_loader = DataLoader.

import torchvisiontrain_set = torchvision.datasets.CIFAR10(root="./dataset",train=True,download=True)test_set = torchvision.datasets.CIFAR10(root="./dataset",train=False,download=True)# train = True 训练集 False 测试集# download = True 下载print(test_set[0]).

输入 输出 作用PIL Image,open()tensor ToTensor()warrays cv.imread()from PIL import Imagefrom torch.utils.tensorboard import SummaryWriterfrom torchvision import transformswriter = SummaryWriter('logs')img = Image.open("data/t...