Skip to content

8~13주차 번역 / 오영화 #51

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

8~13주차 번역 / 오영화 #51

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
180 changes: 90 additions & 90 deletions docs/ko/diffusion/stable_diffusion/sampler/ddim.html
View file
Open in desktop

Large diffs are not rendered by default.

24 changes: 12 additions & 12 deletions docs/ko/gan/wasserstein/gradient_penalty/readme.html
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,26 +1,26 @@
<!DOCTYPE html>
<html lang="en">
<html lang="ko">
<head>
<meta http-equiv="content-type" content="text/html;charset=utf-8"/>
<meta name="viewport" content="width=device-width, initial-scale=1.0"/>
<meta name="description" content=""/>

<meta name="twitter:card" content="summary"/>
<meta name="twitter:card" content="요약"/>
<meta name="twitter:image:src" content="https://avatars1.githubusercontent.com/u/64068543?s=400&amp;v=4"/>
<meta name="twitter:title" content="Gradient Penalty for Wasserstein GAN (WGAN-GP)"/>
<meta name="twitter:title" content="Wasserstein GAN(WGAN-GP)에 대한 GAN 패널티"/>
<meta name="twitter:description" content=""/>
<meta name="twitter:site" content="@labmlai"/>
<meta name="twitter:creator" content="@labmlai"/>

<meta property="og:url" content="https://nn.labml.ai/gan/wasserstein/gradient_penalty/readme.html"/>
<meta property="og:title" content="Gradient Penalty for Wasserstein GAN (WGAN-GP)"/>
<meta property="og:title" content="Wasserstein GAN(WGAN-GP)에 대한 GAN 패널티"/>
<meta property="og:image" content="https://avatars1.githubusercontent.com/u/64068543?s=400&amp;v=4"/>
<meta property="og:site_name" content="Gradient Penalty for Wasserstein GAN (WGAN-GP)"/>
<meta property="og:site_name" content="Wasserstein GAN(WGAN-GP)에 대한 GAN 패널티"/>
<meta property="og:type" content="object"/>
<meta property="og:title" content="Gradient Penalty for Wasserstein GAN (WGAN-GP)"/>
<meta property="og:title" content="Wasserstein GAN(WGAN-GP)에 대한 GAN 패널티"/>
<meta property="og:description" content=""/>

<title>Gradient Penalty for Wasserstein GAN (WGAN-GP)</title>
<title>Wasserstein GAN(WGAN-GP)에 대한 GAN 패널티</title>
<link rel="shortcut icon" href="/icon.png"/>
<link rel="stylesheet" href="../../../pylit.css?v=1">
<link rel="canonical" href="https://nn.labml.ai/gan/wasserstein/gradient_penalty/readme.html"/>
Expand Down Expand Up @@ -72,11 +72,11 @@
<div class='section-link'>
<a href='#section-0'>#</a>
</div>
<h1><a href="https://nn.labml.ai/gan/wasserstein/gradient_penalty/index.html">Gradient Penalty for Wasserstein GAN (WGAN-GP)</a></h1>
<p>This is an implementation of <a href="https://papers.labml.ai/paper/1704.00028">Improved Training of Wasserstein GANs</a>.</p>
<p><a href="https://nn.labml.ai/gan/wasserstein/index.html">WGAN</a> suggests clipping weights to enforce Lipschitz constraint on the discriminator network (critic). This and other weight constraints like L2 norm clipping, weight normalization, L1, L2 weight decay have problems:</p>
<p>1. Limiting the capacity of the discriminator 2. Exploding and vanishing gradients (without <a href="https://nn.labml.ai/normalization/batch_norm/index.html">Batch Normalization</a>).</p>
<p>The paper <a href="https://papers.labml.ai/paper/1704.00028">Improved Training of Wasserstein GANs</a> proposal a better way to improve Lipschitz constraint, a gradient penalty. </p>
<h1><a href="https://nn.labml.ai/gan/wasserstein/gradient_penalty/index.html">Wasserstein GAN(WGAN-GP)에 대한 GAN 패널티</a></h1>
<p>This is an implementation of <a href="https://papers.labml.ai/paper/1704.00028">Wasserstein GANs의 향상된 훈련 방법</a> 논문에 대한 구현입니다.</p>
<p><a href="https://nn.labml.ai/gan/wasserstein/index.html">WGAN</a> 는 판별기 네트워크(critic)에서 립시츠 제약을 적용하기 위해 가중치를 클리핑할것을 제안합니다. 이와 L2 정규 클리핑, 가중치 정규화, L1, L2 가중치 감소와 같은 다른 가중치 제약은 문제가 있습니다 : </p>
<p>1. 판별기의 용량 제한 2. 그래디언트의 폭발 및 소멸 (<a href="https://nn.labml.ai/normalization/batch_norm/index.html">배치 정규화</a>없이).</p>
<p> <a href="https://papers.labml.ai/paper/1704.00028">Wasserstein GANs의 향상된 훈련 방법 </a> 논문은 기울기 패널티인 립시츠 제약을 개선하는 더 나은 방법을 제안합니다. </p>

</div>
<div class='code'>
Expand Down
6 changes: 3 additions & 3 deletions docs/ko/gan/wasserstein/readme.html
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
<!DOCTYPE html>
<html lang="en">
<html lang="ko">
<head>
<meta http-equiv="content-type" content="text/html;charset=utf-8"/>
<meta name="viewport" content="width=device-width, initial-scale=1.0"/>
Expand Down Expand Up @@ -62,7 +62,7 @@
</p>
<p>
<a href="https://github.com/labmlai/annotated_deep_learning_paper_implementations/tree/master/labml_nn/gan/wasserstein/readme.md" target="_blank">
View code on Github</a>
깃헙에서 코드보기</a>
</p>
</div>
</div>
Expand All @@ -72,7 +72,7 @@
<a href='#section-0'>#</a>
</div>
<h1><a href="https://nn.labml.ai/gan/wasserstein/index.html">Wasserstein GAN - WGAN</a></h1>
<p>This is an implementation of <a href="https://papers.labml.ai/paper/1701.07875">Wasserstein GAN</a>. </p>
<p><a href="https://papers.labml.ai/paper/1701.07875">Wasserstein GAN</a>논문의 구현입니다. </p>

</div>
<div class='code'>
Expand Down
32 changes: 16 additions & 16 deletions docs/ko/hypernetworks/experiment.html
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
<!DOCTYPE html>
<html lang="en">
<html lang="ko">
<head>
<meta http-equiv="content-type" content="text/html;charset=utf-8"/>
<meta name="viewport" content="width=device-width, initial-scale=1.0"/>
Expand Down Expand Up @@ -61,7 +61,7 @@
</p>
<p>
<a href="https://github.com/labmlai/annotated_deep_learning_paper_implementations/tree/master/labml_nn/hypernetworks/experiment.py" target="_blank">
View code on Github</a>
깃헙에서 코드보기</a>
</p>
</div>
</div>
Expand Down Expand Up @@ -90,7 +90,7 @@
<div class='section-link'>
<a href='#section-1'>#</a>
</div>
<h2>Auto regressive model</h2>
<h2>자동 회귀 모형</h2>

</div>
<div class='code'>
Expand All @@ -114,7 +114,7 @@ <h2>Auto regressive model</h2>
<div class='section-link'>
<a href='#section-3'>#</a>
</div>
<p>Token embedding module </p>
<p>토큰 임베딩 모듈 </p>

</div>
<div class='code'>
Expand All @@ -140,8 +140,8 @@ <h2>Auto regressive model</h2>
<div class='section-link'>
<a href='#section-5'>#</a>
</div>
<p>Embed the tokens (<code class="highlight"><span></span><span class="n">src</span></code>
) and run it through the the transformer </p>
<p>토큰 (<code class="highlight"><span></span><span class="n">src</span></code>
)을 내장하고 트랜스포머를 통해 실행합니다. </p>

</div>
<div class='code'>
Expand All @@ -153,7 +153,7 @@ <h2>Auto regressive model</h2>
<div class='section-link'>
<a href='#section-6'>#</a>
</div>
<p>Generate logits of the next token </p>
<p>다음 토큰의 로짓(logit)을 생성합니다. </p>

</div>
<div class='code'>
Expand All @@ -165,8 +165,8 @@ <h2>Auto regressive model</h2>
<div class='section-link'>
<a href='#section-7'>#</a>
</div>
<h2>Configurations</h2>
<p>The default configs can and will be over-ridden when we start the experiment</p>
<h2>설정값</h2>
<p>기본 설정은 실험을 시작할 때 재정의될 수 있고 또한 재정의될 것입니다.</p>

</div>
<div class='code'>
Expand Down Expand Up @@ -194,7 +194,7 @@ <h2>Configurations</h2>
<div class='section-link'>
<a href='#section-9'>#</a>
</div>
<p> Initialize the auto-regressive model</p>
<p> 자동 회귀 모델 을 초기화합니다</p>

</div>
<div class='code'>
Expand Down Expand Up @@ -240,7 +240,7 @@ <h2>Configurations</h2>
<div class='section-link'>
<a href='#section-12'>#</a>
</div>
<p>Create experiment </p>
<p>실험을 생성합니다 </p>

</div>
<div class='code'>
Expand All @@ -252,7 +252,7 @@ <h2>Configurations</h2>
<div class='section-link'>
<a href='#section-13'>#</a>
</div>
<p>Create configs </p>
<p>설정값을 생성합니다 </p>

</div>
<div class='code'>
Expand All @@ -264,7 +264,7 @@ <h2>Configurations</h2>
<div class='section-link'>
<a href='#section-14'>#</a>
</div>
<p>Load configurations </p>
<p>설정값을 불러옵니다 </p>

</div>
<div class='code'>
Expand All @@ -276,7 +276,7 @@ <h2>Configurations</h2>
<div class='section-link'>
<a href='#section-15'>#</a>
</div>
<p>A dictionary of configurations to override </p>
<p>재설정하기 위한 설정값이 담긴 딕셔너리 입니다. </p>

</div>
<div class='code'>
Expand All @@ -303,7 +303,7 @@ <h2>Configurations</h2>
<div class='section-link'>
<a href='#section-16'>#</a>
</div>
<p>Set models for saving and loading </p>
<p>모델 저장과 불러오기를 위해 설정합니다. </p>

</div>
<div class='code'>
Expand All @@ -315,7 +315,7 @@ <h2>Configurations</h2>
<div class='section-link'>
<a href='#section-17'>#</a>
</div>
<p>Start the experiment </p>
<p>실험을 시작합니다 </p>

</div>
<div class='code'>
Expand Down
Loading