\n \uc778\uacf5\uc9c0\ub2a5\uacfc \uba38\uc2e0\ub7ec\ub2dd\uc758 \ubc1c\uc804\uc740 \uc6b0\ub9ac\uac00 \uc0dd\ud65c\ud558\uace0 \uc788\ub294 \ubaa8\ub4e0 \ubd84\uc57c\uc5d0 \ud601\uc2e0\uc744 \uac00\uc838\uc654\uc2b5\ub2c8\ub2e4. \uadf8 \uc911\uc5d0\uc11c\ub3c4 GAN(Generative Adversarial Networks)\uacfc RNN(Recurrent Neural Networks)\uc740 \ub9e4\uc6b0 \uac15\ub825\ud55c \ub525\ub7ec\ub2dd \uae30\ubc95\uc73c\ub85c \uc8fc\ubaa9\ubc1b\uace0 \uc788\uc2b5\ub2c8\ub2e4.
\n \uc774 \uae00\uc5d0\uc11c\ub294 \ud30c\uc774\ud1a0\uce58(PyTorch)\ub97c \ud65c\uc6a9\ud558\uc5ec GAN \ubaa8\ub378\uc744 \uad6c\ud604\ud558\uace0, RNN\uc758 \ud6c8\ub828 \ub370\uc774\ud130 \uc218\uc9d1 \ubc29\ubc95\uc5d0 \ub300\ud574 \uc790\uc138\ud788 \ub2e4\ub8e8\uaca0\uc2b5\ub2c8\ub2e4.\n <\/p>\n
\n GAN\uc740 \ub450 \uac1c\uc758 \uc2e0\uacbd\ub9dd(Generator\uc640 Discriminator)\uc774 \uc11c\ub85c \uacbd\uc7c1\ud558\uc5ec \ud559\uc2b5\ud558\ub294 \ubc29\uc2dd\uc785\ub2c8\ub2e4.
\n \uc81c\ub108\ub808\uc774\ud130(Generator)\ub294 \ud604\uc2e4\uacfc \uc720\uc0ac\ud55c \ub370\uc774\ud130\ub97c \uc0dd\uc131\ud558\uace0, \ud310\ubcc4\uae30(Discriminator)\ub294 \uc774 \ub370\uc774\ud130\uac00 \uc9c4\uc9dc\uc778\uc9c0 \uac00\uc9dc\uc778\uc9c0\ub97c \ud310\ub2e8\ud569\ub2c8\ub2e4.
\n GAN\uc740 \uc774\ubbf8\uc9c0 \uc0dd\uc131, \ube44\ub514\uc624 \uc0dd\uc131, \uc74c\uc545 \uc0dd\uc131 \ub4f1 \ub2e4\uc591\ud55c \ubd84\uc57c\uc5d0\uc11c \ud65c\uc6a9\ub418\uace0 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n
\n GAN\uc740 \ub450 \ubd80\ubd84\uc73c\ub85c \ub098\ub258\uc5b4 \uc788\uc2b5\ub2c8\ub2e4:\n <\/p>\n
\n \ub450 \ub124\ud2b8\uc6cc\ud06c\ub294 \uc11c\ub85c\uc758 \uc131\ub2a5\uc744 \ud5a5\uc0c1\uc2dc\ud0a4\uae30 \uc704\ud574 \uacbd\uc7c1\ud558\uba70, \uc774 \uacfc\uc815\uc744 \ud1b5\ud574 \ub354 \ub192\uc740 \ud488\uc9c8\uc758 \ub370\uc774\ud130\ub97c \uc0dd\uc131\ud558\uac8c \ub429\ub2c8\ub2e4.\n <\/p>\n
\n GAN\uc758 \ud559\uc2b5 \uacfc\uc815\uc740 \uc77c\ubc18\uc801\uc73c\ub85c \ub2e4\uc74c \ub2e8\uacc4\ub97c \ud3ec\ud568\ud569\ub2c8\ub2e4:\n <\/p>\n
\n \uba3c\uc800, \ud30c\uc774\ud1a0\uce58 \ub77c\uc774\ube0c\ub7ec\ub9ac\ub97c \uc124\uce58\ud574\uc57c \ud569\ub2c8\ub2e4. \uc544\ub798\uc758 \uba85\ub839\uc5b4\ub97c \uc2e4\ud589\ud558\uc5ec \uc124\uce58\ud558\uc138\uc694.\n <\/p>\n
pip install torch torchvision<\/code><\/pre>\n\ud30c\uc774\ud1a0\uce58\ub97c \ud65c\uc6a9\ud55c GAN \ucf54\ub4dc \uc608\uc81c<\/h3>\n
\n \uc544\ub798\ub294 \uac04\ub2e8\ud55c GAN\uc758 \uad6c\ud604 \uc608\uc81c\uc785\ub2c8\ub2e4. MNIST \ub370\uc774\ud130\uc14b\uc744 \uc0ac\uc6a9\ud558\uc5ec \uc190\uae00\uc528 \uc22b\uc790\ub97c \uc0dd\uc131\ud558\ub294 \ubaa8\ub378\uc744 \ub9cc\ub4e4\uc5b4 \ubcf4\uaca0\uc2b5\ub2c8\ub2e4.\n <\/p>\n
\nimport torch\nimport torch.nn as nn\nimport torch.optim as optim\nimport torchvision.transforms as transforms\nfrom torchvision import datasets\nfrom torch.utils.data import DataLoader\n\n# \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130\nlatent_size = 64\nbatch_size = 100\nlearning_rate = 0.0002\nnum_epochs = 200\n\n# \ub370\uc774\ud130\uc14b \ub85c\ub4dc\ntransform = transforms.Compose([\n transforms.ToTensor(),\n transforms.Normalize((0.5,), (0.5,))\n])\n\nmnist = datasets.MNIST(root='.\/data', train=True, transform=transform, download=True)\ndata_loader = DataLoader(dataset=mnist, batch_size=batch_size, shuffle=True)\n\n# \uc0dd\uc131\uc790 \ud074\ub798\uc2a4 \uc815\uc758\nclass Generator(nn.Module):\n def __init__(self):\n super(Generator, self).__init__()\n self.main = nn.Sequential(\n nn.Linear(latent_size, 256),\n nn.ReLU(True),\n nn.Linear(256, 512),\n nn.ReLU(True),\n nn.Linear(512, 1024),\n nn.ReLU(True),\n nn.Linear(1024, 784),\n nn.Tanh()\n )\n\n def forward(self, x):\n return self.main(x)\n\n# \ud310\ubcc4\uc790 \ud074\ub798\uc2a4 \uc815\uc758\nclass Discriminator(nn.Module):\n def __init__(self):\n super(Discriminator, self).__init__()\n self.main = nn.Sequential(\n nn.Linear(784, 1024),\n nn.LeakyReLU(0.2, inplace=True),\n nn.Linear(1024, 512),\n nn.LeakyReLU(0.2, inplace=True),\n nn.Linear(512, 256),\n nn.LeakyReLU(0.2, inplace=True),\n nn.Linear(256, 1),\n nn.Sigmoid()\n )\n\n def forward(self, x):\n return self.main(x)\n\ngenerator = Generator().cuda()\ndiscriminator = Discriminator().cuda()\n\ncriterion = nn.BCELoss()\noptimizer_G = optim.Adam(generator.parameters(), lr=learning_rate)\noptimizer_D = optim.Adam(discriminator.parameters(), lr=learning_rate)\n\n# \ud559\uc2b5 \uc2dc\uc791\nfor epoch in range(num_epochs):\n for i, (images, _) in enumerate(data_loader):\n # \uc9c4\uc9dc \ub370\uc774\ud130 \ub808\uc774\ube14\n real_images = images.view(-1, 28*28).cuda()\n real_labels = torch.ones(batch_size, 1).cuda()\n # \uac00\uc9dc \ub370\uc774\ud130 \ub808\uc774\ube14\n noise = torch.randn(batch_size, latent_size).cuda()\n fake_images = generator(noise)\n fake_labels = torch.zeros(batch_size, 1).cuda()\n\n # \ud310\ubcc4\uae30 \ud559\uc2b5\n optimizer_D.zero_grad()\n outputs_real = discriminator(real_images)\n outputs_fake = discriminator(fake_images.detach())\n loss_D_real = criterion(outputs_real, real_labels)\n loss_D_fake = criterion(outputs_fake, fake_labels)\n loss_D = loss_D_real + loss_D_fake\n loss_D.backward()\n optimizer_D.step()\n\n # \uc0dd\uc131\uc790 \ud559\uc2b5\n optimizer_G.zero_grad()\n outputs = discriminator(fake_images)\n loss_G = criterion(outputs, real_labels)\n loss_G.backward()\n optimizer_G.step()\n\n print(f\"Epoch [{epoch+1}\/{num_epochs}], Loss D: {loss_D.item()}, Loss G: {loss_G.item()}\")\n if (epoch+1) % 10 == 0:\n # \uacb0\uacfc \uc800\uc7a5 \ucf54\ub4dc \ucd94\uac00 \uac00\ub2a5\n pass\n <\/code><\/pre>\n5. RNN(Recurrent Neural Network) \uc18c\uac1c<\/h2>\n
\n RNN\uc740 \uc21c\uc11c\uac00 \uc788\ub294 \ub370\uc774\ud130, \uc989 \uc2dc\ud000\uc2a4 \ub370\uc774\ud130\ub97c \ucc98\ub9ac\ud558\ub294 \ub370 \uc801\ud569\ud55c \uc2e0\uacbd\ub9dd \uad6c\uc870\uc785\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4, \ud14d\uc2a4\ud2b8, \uc74c\uc545, \uc2dc\uac04 \ub370\uc774\ud130\uc640 \uac19\uc740 \ud615\ud0dc\uc758 \ub370\uc774\ud130\uac00 \uc774\uc5d0 \ud574\ub2f9\ud569\ub2c8\ub2e4.
\n RNN\uc740 \uc774\uc804 \uc0c1\ud0dc\ub97c \uae30\uc5b5\ud558\uace0 \uc774\ub97c \uae30\ubc18\uc73c\ub85c \ud604\uc7ac \uc0c1\ud0dc\ub97c \uc5c5\ub370\uc774\ud2b8\ud558\uc5ec \ub370\uc774\ud130\ub97c \ucc98\ub9ac\ud558\ub294 \ubc29\uc2dd\uc73c\ub85c \uc791\ub3d9\ud569\ub2c8\ub2e4.\n <\/p>\n
RNN\uc758 \uad6c\uc870<\/h3>\n
\n RNN\uc740 \ub2e4\uc74c\uacfc \uac19\uc740 \uad6c\uc131 \uc694\uc18c\ub85c \uc774\ub8e8\uc5b4\uc838 \uc788\uc2b5\ub2c8\ub2e4:\n <\/p>\n
\n- \uc785\ub825\uce35<\/strong>: \ubaa8\ub378\uc758 \uccab \ubc88\uc9f8 \uce35\uc73c\ub85c, \uc2dc\ud000\uc2a4 \ub370\uc774\ud130\ub97c \uc785\ub825\ubc1b\uc2b5\ub2c8\ub2e4.<\/li>\n
- \uc740\ub2c9\uce35<\/strong>: \uc774\uc804 \uc0c1\ud0dc\ub97c \uae30\uc5b5\ud558\uace0 \ud604\uc7ac \uc785\ub825\uc744 \uacb0\ud569\ud558\uc5ec \ucd9c\ub825\uc744 \uc0dd\uc131\ud569\ub2c8\ub2e4.<\/li>\n
- \ucd9c\ub825\uce35<\/strong>: \ucd5c\uc885 \ucd9c\ub825\uc744 \uc0dd\uc131\ud558\ub294 \uce35\uc785\ub2c8\ub2e4.<\/li>\n<\/ul>\n
6. RNN\uc758 \ud6c8\ub828 \ub370\uc774\ud130 \uc218\uc9d1<\/h2>\n
\n RNN\uc744 \ud6c8\ub828\uc2dc\ud0a4\uae30 \uc704\ud574\uc11c\ub294 \uc801\uc808\ud55c \ud6c8\ub828 \ub370\uc774\ud130\uac00 \ud544\uc694\ud569\ub2c8\ub2e4. \uc5ec\uae30\uc11c\ub294 \ud14d\uc2a4\ud2b8 \ub370\uc774\ud130\ub97c \uc218\uc9d1\ud558\uace0 \uc804\ucc98\ub9ac\ud558\ub294 \uacfc\uc815\uc744 \uc124\uba85\ud558\uaca0\uc2b5\ub2c8\ub2e4.\n <\/p>\n
6.1 \ub370\uc774\ud130 \uc218\uc9d1<\/h3>\n
\n RNN\uc758 \ud6c8\ub828\uc5d0 \uc0ac\uc6a9\ud560 \uc218 \uc788\ub294 \ub370\uc774\ud130\ub294 \ub2e4\uc591\ud569\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4, \uc601\ud654 \ub9ac\ubdf0, \uc18c\uc124, \ub274\uc2a4 \uae30\uc0ac \ub4f1 \uc5ec\ub7ec \ud615\ud0dc\uc758 \ud14d\uc2a4\ud2b8 \ub370\uc774\ud130\uac00 \uac00\ub2a5\ud569\ub2c8\ub2e4.
\n \uc6f9 \uc2a4\ud06c\ub798\ud551 \ub3c4\uad6c(\uc608: BeautifulSoup)\ub97c \uc0ac\uc6a9\ud558\uc5ec \ub370\uc774\ud130\ub97c \uc218\uc9d1\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n
\nimport requests\nfrom bs4 import BeautifulSoup\n\nurl = 'https:\/\/example.com\/articles' # \uc6d0\ud558\ub294 URL\ub85c \ubcc0\uacbd\nresponse = requests.get(url)\nsoup = BeautifulSoup(response.content, 'html.parser')\n\narticles = []\nfor item in soup.find_all('article'):\n title = item.find('h2').text\n content = item.find('p').text\n articles.append(f\"{title}\\n{content}\")\n\nwith open('data.txt', 'w', encoding='utf-8') as f:\n for article in articles:\n f.write(article + \"\\n\\n\")\n <\/code><\/pre>\n6.2 \ub370\uc774\ud130 \uc804\ucc98\ub9ac<\/h3>\n
\n \uc218\uc9d1\ud55c \ub370\uc774\ud130\ub294 RNN \ubaa8\ub378\uc758 \uc785\ub825\uc73c\ub85c \uc0ac\uc6a9\ud558\uae30 \uc804, \uc804\ucc98\ub9ac \uacfc\uc815\uc744 \uac70\uccd0\uc57c \ud569\ub2c8\ub2e4. \uc77c\ubc18\uc801\uc778 \uc804\ucc98\ub9ac \uacfc\uc815\uc740 \ub2e4\uc74c\uacfc \uac19\uc2b5\ub2c8\ub2e4:\n <\/p>\n
\n- \uc18c\ubb38\uc790\ud654<\/li>\n
- \ud2b9\uc218\ubb38\uc790 \ubc0f \uc22b\uc790 \uc81c\uac70<\/li>\n
- \ubd88\uc6a9\uc5b4 \uc81c\uac70<\/li>\n<\/ul>\n
\nimport re\nimport nltk\nfrom nltk.corpus import stopwords\n\n# nltk\uc758 \ubd88\uc6a9\uc5b4 \ubaa9\ub85d \ub2e4\uc6b4\ub85c\ub4dc\nnltk.download('stopwords')\nstop_words = set(stopwords.words('english'))\n\ndef preprocess_text(text):\n # \uc18c\ubb38\uc790\ud654\n text = text.lower()\n # \ud2b9\uc218\ubb38\uc790 \ubc0f \uc22b\uc790 \uc81c\uac70\n text = re.sub(r'[^a-z\\s]', '', text)\n # \ubd88\uc6a9\uc5b4 \uc81c\uac70\n text = ' '.join([word for word in text.split() if word not in stop_words])\n return text\n\n# \uc804\ucc98\ub9ac \uc801\uc6a9\npreprocessed_articles = [preprocess_text(article) for article in articles]\n <\/code><\/pre>\n7. RNN \ubaa8\ub378 \uad6c\ud604 \uc608\uc81c<\/h2>\n\ud658\uacbd \uc124\uc815<\/h3>\npip install torch torchvision nltk<\/code><\/pre>\n\ud30c\uc774\ud1a0\uce58\ub97c \ud65c\uc6a9\ud55c RNN \ucf54\ub4dc \uc608\uc81c<\/h3>\n
\n \uc544\ub798\ub294 \uac04\ub2e8\ud55c RNN \ubaa8\ub378\uc758 \uad6c\ud604 \uc608\uc81c\uc785\ub2c8\ub2e4. word embedding\uc744 \uc0ac\uc6a9\ud558\uc5ec \ud14d\uc2a4\ud2b8 \ub370\uc774\ud130\ub97c \ucc98\ub9ac\ud569\ub2c8\ub2e4.\n <\/p>\n
\nimport torch\nimport torch.nn as nn\nimport torch.optim as optim\nfrom torch.utils.data import Dataset, DataLoader\n\n# RNN \ubaa8\ub378 \uc815\uc758\nclass RNN(nn.Module):\n def __init__(self, input_size, hidden_size, output_size):\n super(RNN, self).__init__()\n self.embedding = nn.Embedding(input_size, hidden_size)\n self.rnn = nn.RNN(hidden_size, hidden_size)\n self.fc = nn.Linear(hidden_size, output_size)\n\n def forward(self, x):\n x = self.embedding(x)\n output, hidden = self.rnn(x)\n output = self.fc(output[-1])\n return output\n\n# \ud6c8\ub828 \ub370\uc774\ud130\uc14b \uc0dd\uc131\nclass TextDataset(Dataset):\n def __init__(self, texts, labels):\n self.texts = texts\n self.labels = labels\n\n def __len__(self):\n return len(self.labels)\n\n def __getitem__(self, idx):\n return torch.tensor(self.texts[idx]), torch.tensor(self.labels[idx])\n\n# \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \uc124\uc815\ninput_size = 1000 # \ub2e8\uc5b4 \uc218\nhidden_size = 128\noutput_size = 2 # \ubd84\ub958\ud560 \ud074\ub798\uc2a4 \uc218 (\uc608: \uae0d\uc815\/\ubd80\uc815)\nnum_epochs = 20\nlearning_rate = 0.001\n\n# \ub370\uc774\ud130 \ub85c\ub4dc \ubc0f \uc804\ucc98\ub9ac\n# \uc5ec\uae30\uc11c\ub294 \uac00\uc0c1\uc758 \ub370\uc774\ud130\ub85c \ub300\uccb4\ud569\ub2c8\ub2e4.\ntexts = [...] # \uc804\ucc98\ub9ac\ub41c \ud14d\uc2a4\ud2b8 \ub370\uc774\ud130\nlabels = [...] # \ud574\ub2f9 \ud074\ub798\uc2a4 \ub808\uc774\ube14\n\ndataset = TextDataset(texts, labels)\ndata_loader = DataLoader(dataset, batch_size=32, shuffle=True)\n\n# \ubaa8\ub378 \ucd08\uae30\ud654\nmodel = RNN(input_size, hidden_size, output_size)\ncriterion = nn.CrossEntropyLoss()\noptimizer = optim.Adam(model.parameters(), lr=learning_rate)\n\n# \ud6c8\ub828 \uc2dc\uc791\nfor epoch in range(num_epochs):\n for texts, labels in data_loader:\n optimizer.zero_grad()\n outputs = model(texts)\n loss = criterion(outputs, labels)\n loss.backward()\n optimizer.step()\n\n print(f\"Epoch [{epoch+1}\/{num_epochs}], Loss: {loss.item()}\")\n <\/code><\/pre>\n8. \ub9c8\ubb34\ub9ac<\/h2>\n
\n \uc774\ubc88 \uae00\uc5d0\uc11c\ub294 \ud30c\uc774\ud1a0\uce58\ub97c \ud65c\uc6a9\ud558\uc5ec GAN \ubc0f RNN\uc758 \uae30\ubcf8 \uc6d0\ub9ac\uc640 \uad6c\ud604 \uc608\uc81c\ub97c \ud559\uc2b5\ud558\uc600\uc2b5\ub2c8\ub2e4.
\n GAN\uc744 \uc0ac\uc6a9\ud558\uc5ec \uc774\ubbf8\uc9c0 \ub370\uc774\ud130\ub97c \uc0dd\uc131\ud558\uace0, RNN\uc758 \uacbd\uc6b0 \ud14d\uc2a4\ud2b8 \ub370\uc774\ud130\ub97c \ucc98\ub9ac\ud558\ub294 \uacfc\uc815\uc744 \uc0b4\ud3b4\ubcf4\uc558\uc2b5\ub2c8\ub2e4.
\n \uc774\ub7ec\ud55c \uae30\uc220\ub4e4\uc740 \uc55e\uc73c\ub85c\ub3c4 \uacc4\uc18d \ubc1c\uc804\ud558\uc5ec \ub354 \ub9ce\uc740 \ubd84\uc57c\uc5d0\uc11c \ud65c\uc6a9\ub420 \uac83\uc785\ub2c8\ub2e4.
\n \uc5ec\ub7ec\ubd84\ub3c4 \uc774\ub7ec\ud55c \uae30\uc220\ub4e4\uc744 \ud65c\uc6a9\ud558\uc5ec \uc0c8\ub85c\uc6b4 \ud504\ub85c\uc81d\ud2b8\ub97c \uc2dc\uc791\ud574 \ubcf4\uc2dc\uae30 \ubc14\ub78d\ub2c8\ub2e4.\n <\/p>\n
<\/body><\/p>\n","protected":false},"excerpt":{"rendered":"