您好,登錄后才能下訂單哦!
本文小編為大家詳細(xì)介紹“怎么用GAN訓(xùn)練自己數(shù)據(jù)生成新的圖片”,內(nèi)容詳細(xì),步驟清晰,細(xì)節(jié)處理妥當(dāng),希望這篇“怎么用GAN訓(xùn)練自己數(shù)據(jù)生成新的圖片”文章能幫助大家解決疑惑,下面跟著小編的思路慢慢深入,一起來學(xué)習(xí)新知識(shí)吧。
# MNIST dataset mnist = datasets.MNIST( root='./data/', train=True, transform=img_transform, download=True) # Data loader dataloader = torch.utils.data.DataLoader( dataset=mnist, batch_size=batch_size, shuffle=True)
可以看到,datasets.MNIST這個(gè)肯定不能用于我們自己的數(shù)據(jù)。我借鑒了原來做二分類的datasets.ImageFolder。
發(fā)現(xiàn)老是報(bào)錯(cuò):
RuntimeError: Found 0 files in subfolders of: E:\Projects\gan\battery\ng
Supported extensions are: .jpg,.jpeg,.png,.ppm,.bmp,.pgm,.tif,.tiff,.webp
后面單步調(diào)試,原來這個(gè)函數(shù)是需要文件夾下面有分類標(biāo)簽的,根據(jù)子文件夾名生成分類標(biāo)簽。
故放棄,只能自己寫了。
下面是參考網(wǎng)上的,寫了個(gè)讀取數(shù)據(jù)的函數(shù):
import numpy as np import torch import os import random from PIL import Image from torch.utils.data import Dataset class myDataset(Dataset): def __init__(self, data_dir, transform): self.data_dir = data_dir self.transform = transform self.img_names = [name for name in list(filter(lambda x: x.endswith(".jpg"), os.listdir(self.data_dir)))] def __getitem__(self, index): path_img = os.path.join(self.data_dir, self.img_names[index]) img = Image.open(path_img).convert('RGB') if self.transform is not None: img = self.transform(img) return img def __len__(self): if len(self.img_names) == 0: raise Exception("\ndata_dir:{} is a empty dir! Please checkout your path to images!".format(self.data_dir)) return len(self.img_names)
解決了讀取數(shù)據(jù)之后,發(fā)現(xiàn)可以訓(xùn)練了,因?yàn)閰⒖兼溄拥腗INIST數(shù)據(jù)都是單通道的,我們大部分圖像都是3通道的,所以我將通道改為3后,發(fā)現(xiàn)判別器那塊老是報(bào)錯(cuò),標(biāo)簽和數(shù)據(jù)不匹配。
RuntimeError: mat1 dim 1 must match mat2 dim 0
后面一查,發(fā)現(xiàn)問題出在這句上面:
for i, (imgs, _) in enumerate(dataloader)
這樣得到的imgs已經(jīng)沒有batch-size的信息了,需要改為這樣:
for i, imgs in enumerate(dataloader):
下面是整個(gè)代碼塊,貼上去記錄下來,以便過段時(shí)間萬一忘了,還有個(gè)看的地方。
import argparse import os import numpy as np import math # import torchvision.transforms as transforms from torchvision.utils import save_image from torch.utils.data import DataLoader from torchvision import datasets, models, transforms from torch.autograd import Variable import torch.nn as nn import torch.nn.functional as F from tools.my_dataset import myDataset import torch os.makedirs("images", exist_ok=True) parser = argparse.ArgumentParser() parser.add_argument("--n_epochs", type=int, default=50, help="number of epochs of training") parser.add_argument("--batch_size", type=int, default=2, help="size of the batches") parser.add_argument("--lr", type=float, default=0.0002, help="adam: learning rate") parser.add_argument("--b1", type=float, default=0.5, help="adam: decay of first order momentum of gradient") parser.add_argument("--b2", type=float, default=0.999, help="adam: decay of first order momentum of gradient") parser.add_argument("--n_cpu", type=int, default=4, help="number of cpu threads to use during batch generation") parser.add_argument("--latent_dim", type=int, default=100, help="dimensionality of the latent space") parser.add_argument("--img_size", type=int, default=128, help="size of each image dimension") parser.add_argument("--channels", type=int, default=3, help="number of image channels") parser.add_argument("--sample_interval", type=int, default=400, help="interval betwen image samples") opt = parser.parse_args() print(opt) img_shape = (opt.channels, opt.img_size, opt.img_size) cuda = True if torch.cuda.is_available() else False print('cuda is',cuda) class Generator(nn.Module): def __init__(self): super(Generator, self).__init__() def block(in_feat, out_feat, normalize=True): layers = [nn.Linear(in_feat, out_feat)] if normalize: layers.append(nn.BatchNorm1d(out_feat, 0.8)) layers.append(nn.LeakyReLU(0.2, inplace=True)) return layers self.model = nn.Sequential( *block(opt.latent_dim, 128, normalize=False), *block(128, 256), *block(256, 512), *block(512, 1024), nn.Linear(1024, int(np.prod(img_shape))), nn.Tanh() ) def forward(self, z): img = self.model(z) img = img.view(img.size(0), *img_shape) return img class Discriminator(nn.Module): def __init__(self): super(Discriminator, self).__init__() self.model = nn.Sequential( nn.Linear(int(np.prod(img_shape)), 512), nn.LeakyReLU(0.2, inplace=True), nn.Linear(512, 256), nn.LeakyReLU(0.2, inplace=True), nn.Linear(256, 1), nn.Sigmoid(), ) def forward(self, img): img_flat = img.view(img.size(0), -1) validity = self.model(img_flat) return validity # Loss function adversarial_loss = torch.nn.BCELoss() # Initialize generator and discriminator generator = Generator() discriminator = Discriminator() if cuda: generator.cuda() discriminator.cuda() adversarial_loss.cuda() # # Configure data loader # os.makedirs("./data/mnist", exist_ok=True) # dataloader = torch.utils.data.DataLoader( # datasets.MNIST( # "./data/mnist", # train=True, # download=True, # transform=transforms.Compose( # [transforms.Resize(opt.img_size), transforms.ToTensor(), transforms.Normalize([0.5], [0.5])] # ), # ), # batch_size=opt.batch_size, # shuffle=True, # ) dataset = r'E:\Projects\gan\battery' ng_directory = os.path.join(dataset, 'ng') ok_directory = os.path.join(dataset, 'ok') image_transforms = { 'ng': transforms.Compose([ transforms.Resize([opt.img_size,opt.img_size]), transforms.ToTensor(), ]), 'ok': transforms.Compose([ transforms.Resize([opt.img_size,opt.img_size]), transforms.ToTensor(), ])} data = { 'ng': myDataset(data_dir=ng_directory, transform=image_transforms['ng']), 'ok': myDataset(data_dir=ok_directory, transform=image_transforms['ok']) } dataloader = DataLoader(data['ng'], batch_size=opt.batch_size, shuffle=True) ng_data_size = len(data['ng']) ok_data_size = len(data['ok']) print('train_size: {:4d} valid_size:{:4d}'.format(ng_data_size, ok_data_size)) # Optimizers optimizer_G = torch.optim.Adam(generator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2)) optimizer_D = torch.optim.Adam(discriminator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2)) Tensor = torch.cuda.FloatTensor if cuda else torch.FloatTensor # ---------- # Training # ---------- for epoch in range(opt.n_epochs): # for i, (imgs, _) in enumerate(dataloader): for i, imgs in enumerate(dataloader): # Adversarial ground truths valid = Variable(Tensor(imgs.size(0), 1).fill_(1.0), requires_grad=False) fake = Variable(Tensor(imgs.size(0), 1).fill_(0.0), requires_grad=False) # Configure input real_imgs = Variable(imgs.type(Tensor)) # ----------------- # Train Generator # ----------------- optimizer_G.zero_grad() # Sample noise as generator input z = Variable(Tensor(np.random.normal(0, 3, (imgs.shape[0], opt.latent_dim)))) # Generate a batch of images gen_imgs = generator(z) # Loss measures generator's ability to fool the discriminator aa = discriminator(gen_imgs) g_loss = adversarial_loss(aa, valid) g_loss.backward() optimizer_G.step() # --------------------- # Train Discriminator # --------------------- optimizer_D.zero_grad() # Measure discriminator's ability to classify real from generated samples bb = discriminator(real_imgs) real_loss = adversarial_loss(bb, valid) # 此處需要注意,detach()是為了截?cái)嗵荻攘?,不?jì)算生成網(wǎng)絡(luò)的損失, # 因?yàn)閐_loss包含了fake_loss,回傳的時(shí)候如果不做處理,默認(rèn)會(huì)計(jì)算generator的梯度, # 而這里只需要計(jì)算判別網(wǎng)絡(luò)的梯度,更新其權(quán)重值,生成網(wǎng)絡(luò)保持不變即可。 fake_loss = adversarial_loss(discriminator(gen_imgs.detach()), fake) d_loss = (real_loss + fake_loss) / 2 d_loss.backward() optimizer_D.step() print( "[Epoch %d/%d] [Batch %d/%d] [D loss: %f] [G loss: %f]" % (epoch, opt.n_epochs, i, len(dataloader), d_loss.item(), g_loss.item()) ) batches_done = epoch * len(dataloader) + i if batches_done % opt.sample_interval == 0: save_image(gen_imgs.data[:25], "images/%d.png" % batches_done, nrow=5, normalize=True)
上面是原始圖片,下面是生成的圖片,從開始的噪聲,到慢慢有點(diǎn)樣子,還沒訓(xùn)練完,由于我的顯卡比較小,GTX1660Ti,6G顯存,所以將原始圖片從800x800壓縮到了128x128,可能影響了效果,沒關(guān)系,后面還可以優(yōu)化,包括將全連接網(wǎng)絡(luò)改為卷積的,圖片設(shè)置大點(diǎn),等等。
讀到這里,這篇“怎么用GAN訓(xùn)練自己數(shù)據(jù)生成新的圖片”文章已經(jīng)介紹完畢,想要掌握這篇文章的知識(shí)點(diǎn)還需要大家自己動(dòng)手實(shí)踐使用過才能領(lǐng)會(huì),如果想了解更多相關(guān)內(nèi)容的文章,歡迎關(guān)注億速云行業(yè)資訊頻道。
免責(zé)聲明:本站發(fā)布的內(nèi)容(圖片、視頻和文字)以原創(chuàng)、轉(zhuǎn)載和分享為主,文章觀點(diǎn)不代表本網(wǎng)站立場,如果涉及侵權(quán)請(qǐng)聯(lián)系站長郵箱:is@yisu.com進(jìn)行舉報(bào),并提供相關(guān)證據(jù),一經(jīng)查實(shí),將立刻刪除涉嫌侵權(quán)內(nèi)容。