diffusers/pr_13813/en/_app/immutable/nodes/74.9d52ed6b.js

import{s as ee,n as te,o as ne}from"../chunks/scheduler.53228c21.js";import{S as oe,i as re,e as m,s as r,c as b,h as ae,a as l,d as n,b as a,f as z,g as T,j as Y,k as H,l as v,m as o,n as C,t as w,o as x,p as y}from"../chunks/index.cac5d66a.js";import{C as se}from"../chunks/CopyLLMTxtMenu.5ac9ab94.js";import{D as A}from"../chunks/Docstring.8a316450.js";import{C as de}from"../chunks/CodeBlock.606cbaf4.js";import{H as K,E as ie}from"../chunks/MermaidChart.svelte_svelte_type_style_lang.92b2cd9d.js";function me(Q){let d,Z,D,E,f,j,c,N,g,X="The model can be loaded with the following code snippet.",W,p,R,u,V,s,h,S,M,O="The Transformer model introduced in Longcat-Image.",q,i,_,F,I,B="The forward method.",k,$,G,J,P;return f=new se({props:{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"}}),c=new K({props:{title:"LongCatImageTransformer2DModel",local:"longcatimagetransformer2dmodel",headingTag:"h1"}}),p=new de({props:{code:"ZnJvbSUyMGRpZmZ1c2VycyUyMGltcG9ydCUyMExvbmdDYXRJbWFnZVRyYW5zZm9ybWVyMkRNb2RlbCUwQSUwQXRyYW5zZm9ybWVyJTIwJTNEJTIwTG9uZ0NhdEltYWdlVHJhbnNmb3JtZXIyRE1vZGVsLmZyb21fcHJldHJhaW5lZCglMjJtZWl0dWFuLWxvbmdjYXQlMkZMb25nQ2F0LUltYWdlJTIwJTIyJTJDJTIwc3ViZm9sZGVyJTNEJTIydHJhbnNmb3JtZXIlMjIlMkMlMjB0b3JjaF9kdHlwZSUzRHRvcmNoLmJmbG9hdDE2KQ==",highlighted:`<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> LongCatImageTransformer2DModel

transformer = LongCatImageTransformer2DModel.from_pretrained(<span class="hljs-string">&quot;meituan-longcat/LongCat-Image &quot;</span>, subfolder=<span class="hljs-string">&quot;transformer&quot;</span>, torch_dtype=torch.bfloat16)`,lang:"python",wrap:!1}}),u=new K({props:{title:"LongCatImageTransformer2DModel",local:"diffusers.LongCatImageTransformer2DModel",headingTag:"h2"}}),h=new A({props:{name:"class diffusers.LongCatImageTransformer2DModel",anchor:"diffusers.LongCatImageTransformer2DModel",parameters:[{name:"patch_size",val:": int = 1"},{name:"in_channels",val:": int = 64"},{name:"num_layers",val:": int = 19"},{name:"num_single_layers",val:": int = 38"},{name:"attention_head_dim",val:": int = 128"},{name:"num_attention_heads",val:": int = 24"},{name:"joint_attention_dim",val:": int = 3584"},{name:"pooled_projection_dim",val:": int = 3584"},{name:"axes_dims_rope",val:": list = [16, 56, 56]"}],source:"https://github.com/huggingface/diffusers/blob/vr_13813/src/diffusers/models/transformers/transformer_longcat_image.py#L397"}}),_=new A({props:{name:"forward",anchor:"diffusers.LongCatImageTransformer2DModel.forward",parameters:[{name:"hidden_states",val:": Tensor"},{name:"encoder_hidden_states",val:": Tensor = None"},{name:"timestep",val:": LongTensor = None"},{name:"img_ids",val:": Tensor = None"},{name:"txt_ids",val:": Tensor = None"},{name:"guidance",val:": Tensor = None"},{name:"return_dict",val:": bool = True"}],parametersDescription:[{anchor:"diffusers.LongCatImageTransformer2DModel.forward.hidden_states",description:`<strong>hidden_states</strong> (<code>torch.FloatTensor</code> of shape <code>(batch size, channel, height, width)</code>) &#x2014;
Input <code>hidden_states</code>.`,name:"hidden_states"},{anchor:"diffusers.LongCatImageTransformer2DModel.forward.encoder_hidden_states",description:`<strong>encoder_hidden_states</strong> (<code>torch.FloatTensor</code> of shape <code>(batch size, sequence_len, embed_dims)</code>) &#x2014;
Conditional embeddings (embeddings computed from the input conditions such as prompts) to use.`,name:"encoder_hidden_states"},{anchor:"diffusers.LongCatImageTransformer2DModel.forward.timestep",description:`<strong>timestep</strong> ( <code>torch.LongTensor</code>) &#x2014;
Used to indicate denoising step.`,name:"timestep"},{anchor:"diffusers.LongCatImageTransformer2DModel.forward.img_ids",description:`<strong>img_ids</strong> (<code>torch.Tensor</code>) &#x2014;
Image position ids used to compute the rotary positional embeddings.`,name:"img_ids"},{anchor:"diffusers.LongCatImageTransformer2DModel.forward.txt_ids",description:`<strong>txt_ids</strong> (<code>torch.Tensor</code>) &#x2014;
Text position ids used to compute the rotary positional embeddings.`,name:"txt_ids"},{anchor:"diffusers.LongCatImageTransformer2DModel.forward.guidance",description:`<strong>guidance</strong> (<code>torch.Tensor</code>, <em>optional</em>) &#x2014;
Guidance scale embedding used for guidance-distilled variants of the model.`,name:"guidance"},{anchor:"diffusers.LongCatImageTransformer2DModel.forward.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) &#x2014;
Whether or not to return a <code>~models.transformer_2d.Transformer2DModelOutput</code> instead of a plain
tuple.`,name:"return_dict"}],source:"https://github.com/huggingface/diffusers/blob/vr_13813/src/diffusers/models/transformers/transformer_longcat_image.py#L466",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


<p>If <code>return_dict</code> is True, an <code>~models.transformer_2d.Transformer2DModelOutput</code> is returned, otherwise a
<code>tuple</code> where the first element is the sample tensor.</p>
`}}),$=new ie({props:{source:"https://github.com/huggingface/diffusers/blob/main/docs/source/en/api/models/longcat_image_transformer2d.md"}}),{c(){d=m("meta"),Z=r(),D=m("p"),E=r(),b(f.$$.fragment),j=r(),b(c.$$.fragment),N=r(),g=m("p"),g.textContent=X,W=r(),b(p.$$.fragment),R=r(),b(u.$$.fragment),V=r(),s=m("div"),b(h.$$.fragment),S=r(),M=m("p"),M.textContent=O,q=r(),i=m("div"),b(_.$$.fragment),F=r(),I=m("p"),I.textContent=B,k=r(),b($.$$.fragment),G=r(),J=m("p"),this.h()},l(e){const t=ae("svelte-u9bgzb",document.head);d=l(t,"META",{name:!0,content:!0}),t.forEach(n),Z=a(e),D=l(e,"P",{}),z(D).forEach(n),E=a(e),T(f.$$.fragment,e),j=a(e),T(c.$$.fragment,e),N=a(e),g=l(e,"P",{"data-svelte-h":!0}),Y(g)!=="svelte-1vuni30"&&(g.textContent=X),W=a(e),T(p.$$.fragment,e),R=a(e),T(u.$$.fragment,e),V=a(e),s=l(e,"DIV",{class:!0});var L=z(s);T(h.$$.fragment,L),S=a(L),M=l(L,"P",{"data-svelte-h":!0}),Y(M)!=="svelte-1ib25s7"&&(M.textContent=O),q=a(L),i=l(L,"DIV",{class:!0});var U=z(i);T(_.$$.fragment,U),F=a(U),I=l(U,"P",{"data-svelte-h":!0}),Y(I)!=="svelte-5tc8bf"&&(I.textContent=B),U.forEach(n),L.forEach(n),k=a(e),T($.$$.fragment,e),G=a(e),J=l(e,"P",{}),z(J).forEach(n),this.h()},h(){H(d,"name","hf:doc:metadata"),H(d,"content",le),H(i,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),H(s,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8")},m(e,t){v(document.head,d),o(e,Z,t),o(e,D,t),o(e,E,t),C(f,e,t),o(e,j,t),C(c,e,t),o(e,N,t),o(e,g,t),o(e,W,t),C(p,e,t),o(e,R,t),C(u,e,t),o(e,V,t),o(e,s,t),C(h,s,null),v(s,S),v(s,M),v(s,q),v(s,i),C(_,i,null),v(i,F),v(i,I),o(e,k,t),C($,e,t),o(e,G,t),o(e,J,t),P=!0},p:te,i(e){P||(w(f.$$.fragment,e),w(c.$$.fragment,e),w(p.$$.fragment,e),w(u.$$.fragment,e),w(h.$$.fragment,e),w(_.$$.fragment,e),w($.$$.fragment,e),P=!0)},o(e){x(f.$$.fragment,e),x(c.$$.fragment,e),x(p.$$.fragment,e),x(u.$$.fragment,e),x(h.$$.fragment,e),x(_.$$.fragment,e),x($.$$.fragment,e),P=!1},d(e){e&&(n(Z),n(D),n(E),n(j),n(N),n(g),n(W),n(R),n(V),n(s),n(k),n(G),n(J)),n(d),y(f,e),y(c,e),y(p,e),y(u,e),y(h),y(_),y($,e)}}}const le='{"title":"LongCatImageTransformer2DModel","local":"longcatimagetransformer2dmodel","sections":[{"title":"LongCatImageTransformer2DModel","local":"diffusers.LongCatImageTransformer2DModel","sections":[],"depth":2}],"depth":1}';function fe(Q){return ne(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class $e extends oe{constructor(d){super(),re(this,d,fe,me,ee,{})}}export{$e as component};