diffusers/pr_7930/en/_app/immutable/nodes/16.2ad07c2d.js

import{s as O,n as R,o as U}from"../chunks/scheduler.8c3d61f6.js";import{S as V,i as B,g as _,s,r as F,A as J,h as p,f as n,c as i,j as L,u as y,x as G,k,y as C,a as o,v as z,d as j,t as P,w as E}from"../chunks/index.589a98e8.js";import{D as K}from"../chunks/Docstring.27406313.js";import{H as I,E as Q}from"../chunks/EditOnGithub.e5a8d9cb.js";function W(S){let a,w,h,T,m,b,l,N='A Transformer model for image-like data from <a href="https://blog.fal.ai/auraflow/" rel="nofollow">AuraFlow</a>.',$,d,M,r,f,H,c,q='A 2D Transformer model as introduced in AuraFlow (<a href="https://blog.fal.ai/auraflow/" rel="nofollow">https://blog.fal.ai/auraflow/</a>).',x,u,v,g,A;return m=new I({props:{title:"AuraFlowTransformer2DModel",local:"auraflowtransformer2dmodel",headingTag:"h1"}}),d=new I({props:{title:"AuraFlowTransformer2DModel",local:"diffusers.AuraFlowTransformer2DModel",headingTag:"h2"}}),f=new K({props:{name:"class diffusers.AuraFlowTransformer2DModel",anchor:"diffusers.AuraFlowTransformer2DModel",parameters:[{name:"sample_size",val:": int = 64"},{name:"patch_size",val:": int = 2"},{name:"in_channels",val:": int = 4"},{name:"num_mmdit_layers",val:": int = 4"},{name:"num_single_dit_layers",val:": int = 32"},{name:"attention_head_dim",val:": int = 256"},{name:"num_attention_heads",val:": int = 12"},{name:"joint_attention_dim",val:": int = 2048"},{name:"caption_projection_dim",val:": int = 3072"},{name:"out_channels",val:": int = 4"},{name:"pos_embed_max_size",val:": int = 1024"}],parametersDescription:[{anchor:"diffusers.AuraFlowTransformer2DModel.sample_size",description:`<strong>sample_size</strong> (<code>int</code>) &#x2014; The width of the latent images. This is fixed during training since
it is used to learn a number of position embeddings.`,name:"sample_size"},{anchor:"diffusers.AuraFlowTransformer2DModel.patch_size",description:"<strong>patch_size</strong> (<code>int</code>) &#x2014; Patch size to turn the input data into small patches.",name:"patch_size"},{anchor:"diffusers.AuraFlowTransformer2DModel.in_channels",description:"<strong>in_channels</strong> (<code>int</code>, <em>optional</em>, defaults to 16) &#x2014; The number of channels in the input.",name:"in_channels"},{anchor:"diffusers.AuraFlowTransformer2DModel.num_mmdit_layers",description:"<strong>num_mmdit_layers</strong> (<code>int</code>, <em>optional</em>, defaults to 4) &#x2014; The number of layers of MMDiT Transformer blocks to use.",name:"num_mmdit_layers"},{anchor:"diffusers.AuraFlowTransformer2DModel.num_single_dit_layers",description:`<strong>num_single_dit_layers</strong> (<code>int</code>, <em>optional</em>, defaults to 4) &#x2014;
The number of layers of Transformer blocks to use. These blocks use concatenated image and text
representations.`,name:"num_single_dit_layers"},{anchor:"diffusers.AuraFlowTransformer2DModel.attention_head_dim",description:"<strong>attention_head_dim</strong> (<code>int</code>, <em>optional</em>, defaults to 64) &#x2014; The number of channels in each head.",name:"attention_head_dim"},{anchor:"diffusers.AuraFlowTransformer2DModel.num_attention_heads",description:"<strong>num_attention_heads</strong> (<code>int</code>, <em>optional</em>, defaults to 18) &#x2014; The number of heads to use for multi-head attention.",name:"num_attention_heads"},{anchor:"diffusers.AuraFlowTransformer2DModel.joint_attention_dim",description:"<strong>joint_attention_dim</strong> (<code>int</code>, <em>optional</em>) &#x2014; The number of <code>encoder_hidden_states</code> dimensions to use.",name:"joint_attention_dim"},{anchor:"diffusers.AuraFlowTransformer2DModel.caption_projection_dim",description:"<strong>caption_projection_dim</strong> (<code>int</code>) &#x2014; Number of dimensions to use when projecting the <code>encoder_hidden_states</code>.",name:"caption_projection_dim"},{anchor:"diffusers.AuraFlowTransformer2DModel.out_channels",description:"<strong>out_channels</strong> (<code>int</code>, defaults to 16) &#x2014; Number of output channels.",name:"out_channels"},{anchor:"diffusers.AuraFlowTransformer2DModel.pos_embed_max_size",description:"<strong>pos_embed_max_size</strong> (<code>int</code>, defaults to 4096) &#x2014; Maximum positions to embed from the image latents.",name:"pos_embed_max_size"}],source:"https://github.com/huggingface/diffusers/blob/vr_7930/src/diffusers/models/transformers/auraflow_transformer_2d.py#L235"}}),u=new Q({props:{source:"https://github.com/huggingface/diffusers/blob/main/docs/source/en/api/models/aura_flow_transformer2d.md"}}),{c(){a=_("meta"),w=s(),h=_("p"),T=s(),F(m.$$.fragment),b=s(),l=_("p"),l.innerHTML=N,$=s(),F(d.$$.fragment),M=s(),r=_("div"),F(f.$$.fragment),H=s(),c=_("p"),c.innerHTML=q,x=s(),F(u.$$.fragment),v=s(),g=_("p"),this.h()},l(e){const t=J("svelte-u9bgzb",document.head);a=p(t,"META",{name:!0,content:!0}),t.forEach(n),w=i(e),h=p(e,"P",{}),L(h).forEach(n),T=i(e),y(m.$$.fragment,e),b=i(e),l=p(e,"P",{"data-svelte-h":!0}),G(l)!=="svelte-1cl4wve"&&(l.innerHTML=N),$=i(e),y(d.$$.fragment,e),M=i(e),r=p(e,"DIV",{class:!0});var D=L(r);y(f.$$.fragment,D),H=i(D),c=p(D,"P",{"data-svelte-h":!0}),G(c)!=="svelte-xp13t2"&&(c.innerHTML=q),D.forEach(n),x=i(e),y(u.$$.fragment,e),v=i(e),g=p(e,"P",{}),L(g).forEach(n),this.h()},h(){k(a,"name","hf:doc:metadata"),k(a,"content",X),k(r,"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){C(document.head,a),o(e,w,t),o(e,h,t),o(e,T,t),z(m,e,t),o(e,b,t),o(e,l,t),o(e,$,t),z(d,e,t),o(e,M,t),o(e,r,t),z(f,r,null),C(r,H),C(r,c),o(e,x,t),z(u,e,t),o(e,v,t),o(e,g,t),A=!0},p:R,i(e){A||(j(m.$$.fragment,e),j(d.$$.fragment,e),j(f.$$.fragment,e),j(u.$$.fragment,e),A=!0)},o(e){P(m.$$.fragment,e),P(d.$$.fragment,e),P(f.$$.fragment,e),P(u.$$.fragment,e),A=!1},d(e){e&&(n(w),n(h),n(T),n(b),n(l),n($),n(M),n(r),n(x),n(v),n(g)),n(a),E(m,e),E(d,e),E(f),E(u,e)}}}const X='{"title":"AuraFlowTransformer2DModel","local":"auraflowtransformer2dmodel","sections":[{"title":"AuraFlowTransformer2DModel","local":"diffusers.AuraFlowTransformer2DModel","sections":[],"depth":2}],"depth":1}';function Y(S){return U(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class oe extends V{constructor(a){super(),B(this,a,Y,W,O,{})}}export{oe as component};