diffusers/v0.26.2/ko/_app/immutable/nodes/24.bc95585c.js

import{s as Qe,n as Ve,o as Ge}from"../chunks/scheduler.6e0d5ff7.js";import{S as Fe,i as ke,g as a,s as i,r as M,E as Se,h as p,f as l,c as s,j as Ne,u as m,x as c,k as pe,y as Le,a as n,v as r,d as u,t as f,w as d}from"../chunks/index.d7c1b260.js";import{C as y}from"../chunks/CodeBlock.09a08494.js";import{H as ct}from"../chunks/Heading.30a009b0.js";function Ee(ce){let o,rt,Mt,ut,w,ft,J,Me='<a href="https://arxiv.org/abs/2211.09800" rel="nofollow">InstructPix2Pix</a>는 text-conditioned diffusion 모델이 한 이미지에 편집을 따를 수 있도록 파인튜닝하는 방법입니다. 이 방법을 사용하여 파인튜닝된 모델은 다음을 입력으로 사용합니다:',dt,U,me='<img src="https://huggingface.co/datasets/diffusers/docs-images/resolve/main/evaluation_diffusion_models/edit-instruction.png" alt="instructpix2pix-inputs" width="600/"/>',yt,_,re="출력은 입력 이미지에 편집 지시가 반영된 “수정된” 이미지입니다:",ot,T,ue='<img src="https://huggingface.co/datasets/diffusers/docs-images/resolve/main/output-gs%407-igs%401-steps%4050.png" alt="instructpix2pix-output" width="600/"/>',Ut,b,fe='<code>train_instruct_pix2pix.py</code> 스크립트(<a href="https://github.com/huggingface/diffusers/blob/main/examples/instruct_pix2pix/train_instruct_pix2pix.py" rel="nofollow">여기</a>에서 찾을 수 있습니다.)는 학습 절차를 설명하고 Stable Diffusion에 적용할 수 있는 방법을 보여줍니다.',Tt,W,de='*** <code>train_instruct_pix2pix.py</code>는 <a href="https://github.com/timothybrooks/instruct-pix2pix" rel="nofollow">원래 구현</a>에 충실하면서 InstructPix2Pix 학습 절차를 구현하고 있지만, <a href="https://huggingface.co/datasets/fusing/instructpix2pix-1000-samples" rel="nofollow">소규모 데이터셋</a>에서만 테스트를 했습니다. 이는 최종 결과에 영향을 끼칠 수 있습니다. 더 나은 결과를 위해, 더 큰 데이터셋에서 더 길게 학습하는 것을 권장합니다. <a href="https://huggingface.co/datasets/timbrooks/instructpix2pix-clip-filtered" rel="nofollow">여기</a>에서 InstructPix2Pix 학습을 위해 큰 데이터셋을 찾을 수 있습니다.',wt,Jt,_t,x,bt,h,Wt,Z,ye="이 스크립트를 실행하기 전에, 라이브러리의 학습 종속성을 설치하세요:",xt,C,oe="<strong>중요</strong>",ht,g,Ue="최신 버전의 예제 스크립트를 성공적으로 실행하기 위해, <strong>원본으로부터 설치</strong>하는 것과 예제 스크립트를 자주 업데이트하고 예제별 요구사항을 설치하기 때문에 최신 상태로 유지하는 것을 권장합니다. 이를 위해, 새로운 가상 환경에서 다음 스텝을 실행하세요:",Zt,j,Ct,R,Te="cd 명령어로 예제 폴더로 이동하세요.",gt,$,jt,I,we="이제 실행하세요.",Rt,X,$t,Y,Je='그리고 <a href="https://github.com/huggingface/accelerate/" rel="nofollow">🤗Accelerate</a> 환경에서 초기화하세요:',It,v,Xt,B,_e="혹은 환경에 대한 질문 없이 기본적인 accelerate 구성을 사용하려면 다음을 실행하세요.",Yt,N,vt,Q,be="혹은 사용 중인 환경이 notebook과 같은 대화형 쉘은 지원하지 않는 경우는 다음 절차를 따라주세요.",Bt,V,Nt,G,Qt,F,We='이전에 언급했듯이, 학습을 위해 <a href="https://huggingface.co/datasets/fusing/instructpix2pix-1000-samples" rel="nofollow">작은 데이터셋</a>을 사용할 것입니다. 그 데이터셋은 InstructPix2Pix 논문에서 사용된 <a href="https://huggingface.co/datasets/timbrooks/instructpix2pix-clip-filtered" rel="nofollow">원래의 데이터셋</a>보다 작은 버전입니다. 자신의 데이터셋을 사용하기 위해, <a href="create_dataset">학습을 위한 데이터셋 만들기</a> 가이드를 참고하세요.',Vt,k,xe='<code>MODEL_NAME</code> 환경 변수(허브 모델 레포지토리 또는 모델 가중치가 포함된 폴더 경로)를 지정하고 <a href="https://huggingface.co/docs/diffusers/en/api/diffusion_pipeline#diffusers.DiffusionPipeline.from_pretrained.pretrained_model_name_or_path" rel="nofollow"><code>pretrained_model_name_or_path</code></a> 인수에 전달합니다. <code>DATASET_ID</code>에 데이터셋 이름을 지정해야 합니다:',Gt,S,Ft,L,he="지금, 학습을 실행할 수 있습니다. 스크립트는 레포지토리의 하위 폴더의 모든 구성요소(<code>feature_extractor</code>, <code>scheduler</code>, <code>text_encoder</code>, <code>unet</code> 등)를 저장합니다.",kt,E,St,H,Ze="추가적으로, 가중치와 바이어스를 학습 과정에 모니터링하여 검증 추론을 수행하는 것을 지원합니다. <code>report_to=&quot;wandb&quot;</code>와 이 기능을 사용할 수 있습니다:",Lt,A,Et,z,Ce="모델 디버깅에 유용한 이 평가 방법 권장합니다. 이를 사용하기 위해 <code>wandb</code>를 설치하는 것을 주목해주세요. <code>pip install wandb</code>로 실행해 <code>wandb</code>를 설치할 수 있습니다.",Ht,D,ge='<a href="https://wandb.ai/sayakpaul/instruct-pix2pix/runs/ctr3kovq" rel="nofollow">여기</a>, 몇 가지 평가 방법과 학습 파라미터를 포함하는 예시를 볼 수 있습니다.',At,P,je="<strong><em>참고: 원본 논문에서, 저자들은 256x256 이미지 해상도로 학습한 모델로 512x512와 같은 더 큰 해상도로 잘 일반화되는 것을 볼 수 있었습니다. 이는 학습에 사용한 큰 데이터셋을 사용했기 때문입니다.</em></strong>",zt,q,Dt,O,Re='<code>accelerate</code>는 원활한 다수의 GPU로 학습을 가능하게 합니다. <code>accelerate</code>로 분산 학습을 실행하는 <a href="https://huggingface.co/docs/accelerate/basic_tutorials/launch" rel="nofollow">여기</a> 설명을 따라 해 주시기 바랍니다. 예시의 명령어 입니다:',Pt,K,qt,tt,Ot,et,$e="일단 학습이 완료되면, 추론 할 수 있습니다:",Kt,lt,te,nt,Ie='학습 스크립트를 사용해 얻은 예시의 모델 레포지토리는 여기 <a href="https://huggingface.co/sayakpaul/instruct-pix2pix" rel="nofollow">sayakpaul/instruct-pix2pix</a>에서 확인할 수 있습니다.',ee,it,Xe="성능을 위한 속도와 품질을 제어하기 위해 세 가지 파라미터를 사용하는 것이 좋습니다:",le,st,Ye="<li><code>num_inference_steps</code></li> <li><code>image_guidance_scale</code></li> <li><code>guidance_scale</code></li>",ne,at,ve='특히, <code>image_guidance_scale</code>와 <code>guidance_scale</code>는 생성된(“수정된”) 이미지에서 큰 영향을 미칠 수 있습니다.(<a href="https://twitter.com/RisingSayak/status/1628392199196151808?s=20" rel="nofollow">여기</a>예시를 참고해주세요.)',ie,pt,Be='만약 InstructPix2Pix 학습 방법을 사용해 몇 가지 흥미로운 방법을 찾고 있다면, 이 블로그 게시물<a href="https://huggingface.co/blog/instruction-tuning-sd" rel="nofollow">Instruction-tuning Stable Diffusion with InstructPix2Pix</a>을 확인해주세요.',se,mt,ae;return w=new ct({props:{title:"InstructPix2Pix",local:"instructpix2pix",headingTag:"h1"}}),x=new ct({props:{title:"PyTorch로 로컬에서 실행하기",local:"pytorch로-로컬에서-실행하기",headingTag:"h2"}}),h=new ct({props:{title:"종속성(dependencies) 설치하기",local:"종속성dependencies-설치하기",headingTag:"h3"}}),j=new y({props:{code:"Z2l0JTIwY2xvbmUlMjBodHRwcyUzQSUyRiUyRmdpdGh1Yi5jb20lMkZodWdnaW5nZmFjZSUyRmRpZmZ1c2VycyUwQWNkJTIwZGlmZnVzZXJzJTBBcGlwJTIwaW5zdGFsbCUyMC1lJTIwLg==",highlighted:`git <span class="hljs-built_in">clone</span> https://github.com/huggingface/diffusers
<span class="hljs-built_in">cd</span> diffusers
pip install -e .`,wrap:!1}}),$=new y({props:{code:"Y2QlMjBleGFtcGxlcyUyRmluc3RydWN0X3BpeDJwaXg=",highlighted:'<span class="hljs-built_in">cd</span> examples/instruct_pix2pix',wrap:!1}}),X=new y({props:{code:"cGlwJTIwaW5zdGFsbCUyMC1yJTIwcmVxdWlyZW1lbnRzLnR4dA==",highlighted:"pip install -r requirements.txt",wrap:!1}}),v=new y({props:{code:"YWNjZWxlcmF0ZSUyMGNvbmZpZw==",highlighted:"accelerate config",wrap:!1}}),N=new y({props:{code:"YWNjZWxlcmF0ZSUyMGNvbmZpZyUyMGRlZmF1bHQ=",highlighted:"accelerate config default",wrap:!1}}),V=new y({props:{code:"ZnJvbSUyMGFjY2VsZXJhdGUudXRpbHMlMjBpbXBvcnQlMjB3cml0ZV9iYXNpY19jb25maWclMEElMEF3cml0ZV9iYXNpY19jb25maWcoKQ==",highlighted:`<span class="hljs-keyword">from</span> accelerate.utils <span class="hljs-keyword">import</span> write_basic_config

write_basic_config()`,wrap:!1}}),G=new ct({props:{title:"예시",local:"예시",headingTag:"h3"}}),S=new y({props:{code:"ZXhwb3J0JTIwTU9ERUxfTkFNRSUzRCUyMnJ1bndheW1sJTJGc3RhYmxlLWRpZmZ1c2lvbi12MS01JTIyJTBBZXhwb3J0JTIwREFUQVNFVF9JRCUzRCUyMmZ1c2luZyUyRmluc3RydWN0cGl4MnBpeC0xMDAwLXNhbXBsZXMlMjI=",highlighted:`<span class="hljs-built_in">export</span> MODEL_NAME=<span class="hljs-string">&quot;runwayml/stable-diffusion-v1-5&quot;</span>
<span class="hljs-built_in">export</span> DATASET_ID=<span class="hljs-string">&quot;fusing/instructpix2pix-1000-samples&quot;</span>`,wrap:!1}}),E=new y({props:{code:"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",highlighted:`accelerate launch --mixed_precision=<span class="hljs-string">&quot;fp16&quot;</span> train_instruct_pix2pix.py \\
    --pretrained_model_name_or_path=<span class="hljs-variable">$MODEL_NAME</span> \\
    --dataset_name=<span class="hljs-variable">$DATASET_ID</span> \\
    --enable_xformers_memory_efficient_attention \\
    --resolution=256 --random_flip \\
    --train_batch_size=4 --gradient_accumulation_steps=4 --gradient_checkpointing \\
    --max_train_steps=15000 \\
    --checkpointing_steps=5000 --checkpoints_total_limit=1 \\
    --learning_rate=5e-05 --max_grad_norm=1 --lr_warmup_steps=0 \\
    --conditioning_dropout_prob=0.05 \\
    --mixed_precision=fp16 \\
    --seed=42 \\
    --push_to_hub`,wrap:!1}}),A=new y({props:{code:"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",highlighted:`accelerate launch --mixed_precision=<span class="hljs-string">&quot;fp16&quot;</span> train_instruct_pix2pix.py \\
    --pretrained_model_name_or_path=<span class="hljs-variable">$MODEL_NAME</span> \\
    --dataset_name=<span class="hljs-variable">$DATASET_ID</span> \\
    --enable_xformers_memory_efficient_attention \\
    --resolution=256 --random_flip \\
    --train_batch_size=4 --gradient_accumulation_steps=4 --gradient_checkpointing \\
    --max_train_steps=15000 \\
    --checkpointing_steps=5000 --checkpoints_total_limit=1 \\
    --learning_rate=5e-05 --max_grad_norm=1 --lr_warmup_steps=0 \\
    --conditioning_dropout_prob=0.05 \\
    --mixed_precision=fp16 \\
    --val_image_url=<span class="hljs-string">&quot;https://hf.co/datasets/diffusers/diffusers-images-docs/resolve/main/mountain.png&quot;</span> \\
    --validation_prompt=<span class="hljs-string">&quot;make the mountains snowy&quot;</span> \\
    --seed=42 \\
    --report_to=wandb \\
    --push_to_hub`,wrap:!1}}),q=new ct({props:{title:"다수의 GPU로 학습하기",local:"다수의-gpu로-학습하기",headingTag:"h2"}}),K=new y({props:{code:"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",highlighted:`accelerate launch --mixed_precision=<span class="hljs-string">&quot;fp16&quot;</span> --multi_gpu train_instruct_pix2pix.py \\
 --pretrained_model_name_or_path=runwayml/stable-diffusion-v1-5 \\
 --dataset_name=sayakpaul/instructpix2pix-1000-samples \\
 --use_ema \\
 --enable_xformers_memory_efficient_attention \\
 --resolution=512 --random_flip \\
 --train_batch_size=4 --gradient_accumulation_steps=4 --gradient_checkpointing \\
 --max_train_steps=15000 \\
 --checkpointing_steps=5000 --checkpoints_total_limit=1 \\
 --learning_rate=5e-05 --lr_warmup_steps=0 \\
 --conditioning_dropout_prob=0.05 \\
 --mixed_precision=fp16 \\
 --seed=42 \\
 --push_to_hub`,wrap:!1}}),tt=new ct({props:{title:"추론하기",local:"추론하기",headingTag:"h2"}}),lt=new y({props:{code:"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",highlighted:`<span class="hljs-keyword">import</span> PIL
<span class="hljs-keyword">import</span> requests
<span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> StableDiffusionInstructPix2PixPipeline

model_id = <span class="hljs-string">&quot;your_model_id&quot;</span>  <span class="hljs-comment"># &lt;- 이를 수정하세요.</span>
pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to(<span class="hljs-string">&quot;cuda&quot;</span>)
generator = torch.Generator(<span class="hljs-string">&quot;cuda&quot;</span>).manual_seed(<span class="hljs-number">0</span>)

url = <span class="hljs-string">&quot;https://huggingface.co/datasets/sayakpaul/sample-datasets/resolve/main/test_pix2pix_4.png&quot;</span>


<span class="hljs-keyword">def</span> <span class="hljs-title function_">download_image</span>(<span class="hljs-params">url</span>):
   image = PIL.Image.<span class="hljs-built_in">open</span>(requests.get(url, stream=<span class="hljs-literal">True</span>).raw)
   image = PIL.ImageOps.exif_transpose(image)
   image = image.convert(<span class="hljs-string">&quot;RGB&quot;</span>)
   <span class="hljs-keyword">return</span> image


image = download_image(url)
prompt = <span class="hljs-string">&quot;wipe out the lake&quot;</span>
num_inference_steps = <span class="hljs-number">20</span>
image_guidance_scale = <span class="hljs-number">1.5</span>
guidance_scale = <span class="hljs-number">10</span>

edited_image = pipe(
   prompt,
   image=image,
   num_inference_steps=num_inference_steps,
   image_guidance_scale=image_guidance_scale,
   guidance_scale=guidance_scale,
   generator=generator,
).images[<span class="hljs-number">0</span>]
edited_image.save(<span class="hljs-string">&quot;edited_image.png&quot;</span>)`,wrap:!1}}),{c(){o=a("meta"),rt=i(),Mt=a("p"),ut=i(),M(w.$$.fragment),ft=i(),J=a("p"),J.innerHTML=Me,dt=i(),U=a("p"),U.innerHTML=me,yt=i(),_=a("p"),_.textContent=re,ot=i(),T=a("p"),T.innerHTML=ue,Ut=i(),b=a("p"),b.innerHTML=fe,Tt=i(),W=a("p"),W.innerHTML=de,wt=i(),Jt=a("hr"),_t=i(),M(x.$$.fragment),bt=i(),M(h.$$.fragment),Wt=i(),Z=a("p"),Z.textContent=ye,xt=i(),C=a("p"),C.innerHTML=oe,ht=i(),g=a("p"),g.innerHTML=Ue,Zt=i(),M(j.$$.fragment),Ct=i(),R=a("p"),R.textContent=Te,gt=i(),M($.$$.fragment),jt=i(),I=a("p"),I.textContent=we,Rt=i(),M(X.$$.fragment),$t=i(),Y=a("p"),Y.innerHTML=Je,It=i(),M(v.$$.fragment),Xt=i(),B=a("p"),B.textContent=_e,Yt=i(),M(N.$$.fragment),vt=i(),Q=a("p"),Q.textContent=be,Bt=i(),M(V.$$.fragment),Nt=i(),M(G.$$.fragment),Qt=i(),F=a("p"),F.innerHTML=We,Vt=i(),k=a("p"),k.innerHTML=xe,Gt=i(),M(S.$$.fragment),Ft=i(),L=a("p"),L.innerHTML=he,kt=i(),M(E.$$.fragment),St=i(),H=a("p"),H.innerHTML=Ze,Lt=i(),M(A.$$.fragment),Et=i(),z=a("p"),z.innerHTML=Ce,Ht=i(),D=a("p"),D.innerHTML=ge,At=i(),P=a("p"),P.innerHTML=je,zt=i(),M(q.$$.fragment),Dt=i(),O=a("p"),O.innerHTML=Re,Pt=i(),M(K.$$.fragment),qt=i(),M(tt.$$.fragment),Ot=i(),et=a("p"),et.textContent=$e,Kt=i(),M(lt.$$.fragment),te=i(),nt=a("p"),nt.innerHTML=Ie,ee=i(),it=a("p"),it.textContent=Xe,le=i(),st=a("ul"),st.innerHTML=Ye,ne=i(),at=a("p"),at.innerHTML=ve,ie=i(),pt=a("p"),pt.innerHTML=Be,se=i(),mt=a("p"),this.h()},l(t){const e=Se("svelte-u9bgzb",document.head);o=p(e,"META",{name:!0,content:!0}),e.forEach(l),rt=s(t),Mt=p(t,"P",{}),Ne(Mt).forEach(l),ut=s(t),m(w.$$.fragment,t),ft=s(t),J=p(t,"P",{"data-svelte-h":!0}),c(J)!=="svelte-buipge"&&(J.innerHTML=Me),dt=s(t),U=p(t,"P",{align:!0,"data-svelte-h":!0}),c(U)!=="svelte-boa587"&&(U.innerHTML=me),yt=s(t),_=p(t,"P",{"data-svelte-h":!0}),c(_)!=="svelte-sx3fp4"&&(_.textContent=re),ot=s(t),T=p(t,"P",{align:!0,"data-svelte-h":!0}),c(T)!=="svelte-5k7lrs"&&(T.innerHTML=ue),Ut=s(t),b=p(t,"P",{"data-svelte-h":!0}),c(b)!=="svelte-1e9auo8"&&(b.innerHTML=fe),Tt=s(t),W=p(t,"P",{"data-svelte-h":!0}),c(W)!=="svelte-1r4gmj9"&&(W.innerHTML=de),wt=s(t),Jt=p(t,"HR",{}),_t=s(t),m(x.$$.fragment,t),bt=s(t),m(h.$$.fragment,t),Wt=s(t),Z=p(t,"P",{"data-svelte-h":!0}),c(Z)!=="svelte-3oxxid"&&(Z.textContent=ye),xt=s(t),C=p(t,"P",{"data-svelte-h":!0}),c(C)!=="svelte-kmsm4y"&&(C.innerHTML=oe),ht=s(t),g=p(t,"P",{"data-svelte-h":!0}),c(g)!=="svelte-7udjz2"&&(g.innerHTML=Ue),Zt=s(t),m(j.$$.fragment,t),Ct=s(t),R=p(t,"P",{"data-svelte-h":!0}),c(R)!=="svelte-ptf992"&&(R.textContent=Te),gt=s(t),m($.$$.fragment,t),jt=s(t),I=p(t,"P",{"data-svelte-h":!0}),c(I)!=="svelte-1tdja71"&&(I.textContent=we),Rt=s(t),m(X.$$.fragment,t),$t=s(t),Y=p(t,"P",{"data-svelte-h":!0}),c(Y)!=="svelte-d5vnul"&&(Y.innerHTML=Je),It=s(t),m(v.$$.fragment,t),Xt=s(t),B=p(t,"P",{"data-svelte-h":!0}),c(B)!=="svelte-md9hwf"&&(B.textContent=_e),Yt=s(t),m(N.$$.fragment,t),vt=s(t),Q=p(t,"P",{"data-svelte-h":!0}),c(Q)!=="svelte-g6s7ey"&&(Q.textContent=be),Bt=s(t),m(V.$$.fragment,t),Nt=s(t),m(G.$$.fragment,t),Qt=s(t),F=p(t,"P",{"data-svelte-h":!0}),c(F)!=="svelte-ninc34"&&(F.innerHTML=We),Vt=s(t),k=p(t,"P",{"data-svelte-h":!0}),c(k)!=="svelte-1qxsp6v"&&(k.innerHTML=xe),Gt=s(t),m(S.$$.fragment,t),Ft=s(t),L=p(t,"P",{"data-svelte-h":!0}),c(L)!=="svelte-xbem7q"&&(L.innerHTML=he),kt=s(t),m(E.$$.fragment,t),St=s(t),H=p(t,"P",{"data-svelte-h":!0}),c(H)!=="svelte-8a508b"&&(H.innerHTML=Ze),Lt=s(t),m(A.$$.fragment,t),Et=s(t),z=p(t,"P",{"data-svelte-h":!0}),c(z)!=="svelte-159u2wd"&&(z.innerHTML=Ce),Ht=s(t),D=p(t,"P",{"data-svelte-h":!0}),c(D)!=="svelte-1hhmng0"&&(D.innerHTML=ge),At=s(t),P=p(t,"P",{"data-svelte-h":!0}),c(P)!=="svelte-onv5fd"&&(P.innerHTML=je),zt=s(t),m(q.$$.fragment,t),Dt=s(t),O=p(t,"P",{"data-svelte-h":!0}),c(O)!=="svelte-1kkgxxt"&&(O.innerHTML=Re),Pt=s(t),m(K.$$.fragment,t),qt=s(t),m(tt.$$.fragment,t),Ot=s(t),et=p(t,"P",{"data-svelte-h":!0}),c(et)!=="svelte-1fgn54t"&&(et.textContent=$e),Kt=s(t),m(lt.$$.fragment,t),te=s(t),nt=p(t,"P",{"data-svelte-h":!0}),c(nt)!=="svelte-jtxuis"&&(nt.innerHTML=Ie),ee=s(t),it=p(t,"P",{"data-svelte-h":!0}),c(it)!=="svelte-r4bxfr"&&(it.textContent=Xe),le=s(t),st=p(t,"UL",{"data-svelte-h":!0}),c(st)!=="svelte-1k7xfoe"&&(st.innerHTML=Ye),ne=s(t),at=p(t,"P",{"data-svelte-h":!0}),c(at)!=="svelte-1u94kli"&&(at.innerHTML=ve),ie=s(t),pt=p(t,"P",{"data-svelte-h":!0}),c(pt)!=="svelte-123fa67"&&(pt.innerHTML=Be),se=s(t),mt=p(t,"P",{}),Ne(mt).forEach(l),this.h()},h(){pe(o,"name","hf:doc:metadata"),pe(o,"content",He),pe(U,"align","center"),pe(T,"align","center")},m(t,e){Le(document.head,o),n(t,rt,e),n(t,Mt,e),n(t,ut,e),r(w,t,e),n(t,ft,e),n(t,J,e),n(t,dt,e),n(t,U,e),n(t,yt,e),n(t,_,e),n(t,ot,e),n(t,T,e),n(t,Ut,e),n(t,b,e),n(t,Tt,e),n(t,W,e),n(t,wt,e),n(t,Jt,e),n(t,_t,e),r(x,t,e),n(t,bt,e),r(h,t,e),n(t,Wt,e),n(t,Z,e),n(t,xt,e),n(t,C,e),n(t,ht,e),n(t,g,e),n(t,Zt,e),r(j,t,e),n(t,Ct,e),n(t,R,e),n(t,gt,e),r($,t,e),n(t,jt,e),n(t,I,e),n(t,Rt,e),r(X,t,e),n(t,$t,e),n(t,Y,e),n(t,It,e),r(v,t,e),n(t,Xt,e),n(t,B,e),n(t,Yt,e),r(N,t,e),n(t,vt,e),n(t,Q,e),n(t,Bt,e),r(V,t,e),n(t,Nt,e),r(G,t,e),n(t,Qt,e),n(t,F,e),n(t,Vt,e),n(t,k,e),n(t,Gt,e),r(S,t,e),n(t,Ft,e),n(t,L,e),n(t,kt,e),r(E,t,e),n(t,St,e),n(t,H,e),n(t,Lt,e),r(A,t,e),n(t,Et,e),n(t,z,e),n(t,Ht,e),n(t,D,e),n(t,At,e),n(t,P,e),n(t,zt,e),r(q,t,e),n(t,Dt,e),n(t,O,e),n(t,Pt,e),r(K,t,e),n(t,qt,e),r(tt,t,e),n(t,Ot,e),n(t,et,e),n(t,Kt,e),r(lt,t,e),n(t,te,e),n(t,nt,e),n(t,ee,e),n(t,it,e),n(t,le,e),n(t,st,e),n(t,ne,e),n(t,at,e),n(t,ie,e),n(t,pt,e),n(t,se,e),n(t,mt,e),ae=!0},p:Ve,i(t){ae||(u(w.$$.fragment,t),u(x.$$.fragment,t),u(h.$$.fragment,t),u(j.$$.fragment,t),u($.$$.fragment,t),u(X.$$.fragment,t),u(v.$$.fragment,t),u(N.$$.fragment,t),u(V.$$.fragment,t),u(G.$$.fragment,t),u(S.$$.fragment,t),u(E.$$.fragment,t),u(A.$$.fragment,t),u(q.$$.fragment,t),u(K.$$.fragment,t),u(tt.$$.fragment,t),u(lt.$$.fragment,t),ae=!0)},o(t){f(w.$$.fragment,t),f(x.$$.fragment,t),f(h.$$.fragment,t),f(j.$$.fragment,t),f($.$$.fragment,t),f(X.$$.fragment,t),f(v.$$.fragment,t),f(N.$$.fragment,t),f(V.$$.fragment,t),f(G.$$.fragment,t),f(S.$$.fragment,t),f(E.$$.fragment,t),f(A.$$.fragment,t),f(q.$$.fragment,t),f(K.$$.fragment,t),f(tt.$$.fragment,t),f(lt.$$.fragment,t),ae=!1},d(t){t&&(l(rt),l(Mt),l(ut),l(ft),l(J),l(dt),l(U),l(yt),l(_),l(ot),l(T),l(Ut),l(b),l(Tt),l(W),l(wt),l(Jt),l(_t),l(bt),l(Wt),l(Z),l(xt),l(C),l(ht),l(g),l(Zt),l(Ct),l(R),l(gt),l(jt),l(I),l(Rt),l($t),l(Y),l(It),l(Xt),l(B),l(Yt),l(vt),l(Q),l(Bt),l(Nt),l(Qt),l(F),l(Vt),l(k),l(Gt),l(Ft),l(L),l(kt),l(St),l(H),l(Lt),l(Et),l(z),l(Ht),l(D),l(At),l(P),l(zt),l(Dt),l(O),l(Pt),l(qt),l(Ot),l(et),l(Kt),l(te),l(nt),l(ee),l(it),l(le),l(st),l(ne),l(at),l(ie),l(pt),l(se),l(mt)),l(o),d(w,t),d(x,t),d(h,t),d(j,t),d($,t),d(X,t),d(v,t),d(N,t),d(V,t),d(G,t),d(S,t),d(E,t),d(A,t),d(q,t),d(K,t),d(tt,t),d(lt,t)}}}const He='{"title":"InstructPix2Pix","local":"instructpix2pix","sections":[{"title":"PyTorch로 로컬에서 실행하기","local":"pytorch로-로컬에서-실행하기","sections":[{"title":"종속성(dependencies) 설치하기","local":"종속성dependencies-설치하기","sections":[],"depth":3},{"title":"예시","local":"예시","sections":[],"depth":3}],"depth":2},{"title":"다수의 GPU로 학습하기","local":"다수의-gpu로-학습하기","sections":[],"depth":2},{"title":"추론하기","local":"추론하기","sections":[],"depth":2}],"depth":1}';function Ae(ce){return Ge(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class Oe extends Fe{constructor(o){super(),ke(this,o,Ae,Ee,Qe,{})}}export{Oe as component};