Buckets:

hf-doc-build
/

doc-dev

Files

xet

hf-doc-build/doc-dev / diffusers /pr_14921 /en /_app /immutable /nodes /46.D9JMhtuH.js

HuggingFaceDocBuilder

21 days ago

download

raw

16.1 kB

import"../chunks/DsnmJJEf.js";import{i as E,h as Z,C as F,H as a,a as j,D as o,E as G,s as V}from"../chunks/BLbO27j4.js";import{p as B,o as X,s as e,f as H,a as A,b as C,c as t,d as D,n as r,r as n}from"../chunks/jDjavuwI.js";const R='{"title":"AutoencoderKLQwenImage21","local":"autoencoderklqwenimage21","sections":[{"title":"AutoencoderKLQwenImage21","local":"diffusers.AutoencoderKLQwenImage21","sections":[],"depth":2},{"title":"AutoencoderKLOutput","local":"diffusers.models.modeling_outputs.AutoencoderKLOutput","sections":[],"depth":2},{"title":"DecoderOutput","local":"diffusers.models.autoencoders.vae.DecoderOutput","sections":[],"depth":2}],"depth":1}';var S=D('<meta name="hf:doc:metadata"/>'),U=D(`<p></p> <!> <!> <p>The 64-channel variational auto-encoder used by Qwen-Image 2.1. It compresses 16x spatially, and its per-channel <code>latents_mean</code> / <code>latents_std</code> are part of the config rather than a single scaling factor.</p> <!> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>A VAE model with KL loss for encoding videos into latents and decoding latent representations into videos.
Introduced in [Qwen Image 2].</p> <p>This model inherits from <a href="/docs/diffusers/pr_14921/en/api/models/overview#diffusers.ModelMixin">ModelMixin</a>. Check the superclass documentation for it’s generic methods implemented
for all models (such as downloading or saving).</p> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Decode a batch of images.</p></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Encode a batch of images into latents.</p></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Enable tiled VAE decoding. When this option is enabled, the VAE will split the input tensor into tiles to
compute decoding and encoding in several steps. This is useful for saving a large amount of memory and to allow
processing larger images.</p></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Decode a batch of images using a tiled decoder.</p></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Encode a batch of images using a tiled encoder.</p></div></div> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Output of AutoencoderKL encoding method.</p></div> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Output of decoding method.</p></div> <!> <p></p>`,1);function ee(K,y){B(y,!1),X(()=>{new URLSearchParams(window.location.search).get("fw")}),E();var g=U();Z("nfqv89",L=>{var I=S();V(I,"content",R),A(L,I)});var f=e(H(g),2);F(f,{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"});var h=e(f,2);a(h,{title:"AutoencoderKLQwenImage21",local:"autoencoderklqwenimage21",headingTag:"h1"});var _=e(h,4);j(_,{code:"aW1wb3J0JTIwdG9yY2glMEFmcm9tJTIwZGlmZnVzZXJzJTIwaW1wb3J0JTIwQXV0b2VuY29kZXJLTFF3ZW5JbWFnZTIxJTBBJTBBdmFlJTIwJTNEJTIwQXV0b2VuY29kZXJLTFF3ZW5JbWFnZTIxLmZyb21fcHJldHJhaW5lZCglMjJRd2VuJTJGUXdlbi1JbWFnZS0yLjElMjIlMkMlMjBzdWJmb2xkZXIlM0QlMjJ2YWUlMjIlMkMlMjBkdHlwZSUzRHRvcmNoLmJmbG9hdDE2KQ==",highlighted:`<span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> AutoencoderKLQwenImage21
vae = AutoencoderKLQwenImage21.from_pretrained(<span class="hljs-string">&quot;Qwen/Qwen-Image-2.1&quot;</span>, subfolder=<span class="hljs-string">&quot;vae&quot;</span>, dtype=torch.bfloat16)`,lang:"python",wrap:!1});var v=e(_,2);a(v,{title:"AutoencoderKLQwenImage21",local:"diffusers.AutoencoderKLQwenImage21",headingTag:"h2"});var d=e(v,2),b=t(d);o(b,{name:"class diffusers.AutoencoderKLQwenImage21",anchor:"diffusers.AutoencoderKLQwenImage21",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L977",parameters:[{name:"base_dim",val:": int = 96"},{name:"decoder_base_dim",val:": int | None = 144"},{name:"z_dim",val:": int = 64"},{name:"dim_mult",val:": list = [1, 2, 4, 8, 8]"},{name:"num_res_blocks",val:": int = 2"},{name:"attn_scales",val:": list = []"},{name:"temperal_downsample",val:": list = [False, True, True, True]"},{name:"dropout",val:": float = 0.0"},{name:"latents_mean",val:": list = [0.5126, 0.7721, -0.0631, 1.3506, -0.7855, -2.1025, -0.3458, 1.3722, 1.8873, -1.7177, -0.651, 0.2732, 0.7562, -0.6163, -1.0277, 3.8363, 2.021, 0.0472, 0.932, 2.0087, 2.4954, -0.1391, -1.4249, 1.8464, -0.5236, 1.2826, 3.7046, -1.3035, 2.7286, -1.4518, -1.9036, -1.9955, -0.0342, -1.0265, -0.7636, 3.0555, 0.0746, -3.0751, -0.1076, 1.7376, -1.0914, -1.9435, -0.2784, -1.368, 0.4809, -0.4433, 0.3764, 0.5729, -2.0595, 1.096, -1.326, -2.0211, -5.0179, 0.5275, 4.0162, 1.8505, 0.3026, 1.9373, 1.4937, 0.2632, 0.5547, -1.7121, -0.1562, 0.0304]"},{name:"latents_std",val:": list = [3.2001, 3.2936, 3.4321, 3.0091, 3.1061, 4.0379, 4.0705, 3.791, 3.0785, 3.65, 3.9308, 3.0904, 2.8778, 3.7675, 3.732, 5.0756, 3.2864, 4.0397, 3.1317, 4.0443, 2.9249, 3.9454, 3.0988, 4.2489, 3.4896, 3.8513, 3.9323, 3.4719, 3.7498, 4.283, 3.5694, 4.2467, 3.9037, 3.2947, 5.077, 3.5075, 3.27, 3.4767, 2.8063, 5.1125, 3.5327, 4.7833, 3.1286, 4.1819, 3.8527, 3.8312, 3.5605, 4.3875, 3.9624, 4.0168, 3.5643, 4.055, 5.5614, 4.2963, 4.408, 3.4959, 3.8747, 3.7608, 3.5735, 3.149, 3.7662, 3.6746, 3.4563, 3.8161]"},{name:"is_residual",val:": bool = True"},{name:"in_channels",val:": int = 4"},{name:"out_channels",val:": int = 4"},{name:"patch_size",val:": int | None = None"},{name:"scale_factor_temporal",val:": int | None = 8"},{name:"scale_factor_spatial",val:": int | None = 16"}]});var s=e(b,6),Q=t(s);o(Q,{name:"decode",anchor:"diffusers.AutoencoderKLQwenImage21.decode",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1336",parameters:[{name:"z",val:": Tensor"},{name:"return_dict",val:": bool = True"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.decode.z",description:"<strong>z</strong> (<code>torch.Tensor</code>) &#x2014; Input batch of latent vectors.",name:"z"},{anchor:"diffusers.AutoencoderKLQwenImage21.decode.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) &#x2014;
Whether to return a <code>~models.vae.DecoderOutput</code> instead of a plain tuple.`,name:"return_dict"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p>If return_dict is True, a <code>~models.vae.DecoderOutput</code> is returned, otherwise a plain <code>tuple</code> is
returned.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><code>~models.vae.DecoderOutput</code> or <code>tuple</code></p>
`}),r(2),n(s);var i=e(s,2),O=t(i);o(O,{name:"encode",anchor:"diffusers.AutoencoderKLQwenImage21.encode",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1276",parameters:[{name:"x",val:": Tensor"},{name:"return_dict",val:": bool = True"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.encode.x",description:"<strong>x</strong> (<code>torch.Tensor</code>) &#x2014; Input batch of images.",name:"x"},{anchor:"diffusers.AutoencoderKLQwenImage21.encode.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) &#x2014;
Whether to return a <code>~models.autoencoder_kl.AutoencoderKLOutput</code> instead of a plain tuple.`,name:"return_dict"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p>The latent representations of the encoded videos. If <code>return_dict</code> is True, a
<code>~models.autoencoder_kl.AutoencoderKLOutput</code> is returned, otherwise a plain <code>tuple</code> is returned.</p>
`}),r(2),n(i);var c=e(i,2),k=t(c);o(k,{name:"enable_tiling",anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1207",parameters:[{name:"tile_sample_min_height",val:": int | None = None"},{name:"tile_sample_min_width",val:": int | None = None"},{name:"tile_sample_stride_height",val:": float | None = None"},{name:"tile_sample_stride_width",val:": float | None = None"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling.tile_sample_min_height",description:`<strong>tile_sample_min_height</strong> (<code>int</code>, <em>optional</em>) &#x2014;
The minimum height required for a sample to be separated into tiles across the height dimension.`,name:"tile_sample_min_height"},{anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling.tile_sample_min_width",description:`<strong>tile_sample_min_width</strong> (<code>int</code>, <em>optional</em>) &#x2014;
The minimum width required for a sample to be separated into tiles across the width dimension.`,name:"tile_sample_min_width"},{anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling.tile_sample_stride_height",description:`<strong>tile_sample_stride_height</strong> (<code>int</code>, <em>optional</em>) &#x2014;
The minimum amount of overlap between two consecutive vertical tiles. This is to ensure that there are
no tiling artifacts produced across the height dimension.`,name:"tile_sample_stride_height"},{anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling.tile_sample_stride_width",description:`<strong>tile_sample_stride_width</strong> (<code>int</code>, <em>optional</em>) &#x2014;
The stride between two consecutive horizontal tiles. This is to ensure that there are no tiling
artifacts produced across the width dimension.`,name:"tile_sample_stride_width"}]}),r(2),n(c);var l=e(c,2),J=t(l);o(J,{name:"forward",anchor:"diffusers.AutoencoderKLQwenImage21.forward",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1532",parameters:[{name:"sample",val:": Tensor"},{name:"sample_posterior",val:": bool = False"},{name:"return_dict",val:": bool = True"},{name:"generator",val:": typing.Optional[torch.Generator] = None"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.forward.sample",description:"<strong>sample</strong> (<code>torch.Tensor</code>) &#x2014; Input sample.",name:"sample"},{anchor:"diffusers.AutoencoderKLQwenImage21.forward.sample_posterior",description:`<strong>sample_posterior</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>False</code>) &#x2014;
Whether to sample from the posterior instead of taking its mode.`,name:"sample_posterior"},{anchor:"diffusers.AutoencoderKLQwenImage21.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>DecoderOutput</code> instead of a plain tuple.`,name:"return_dict"},{anchor:"diffusers.AutoencoderKLQwenImage21.forward.generator",description:`<strong>generator</strong> (<code>torch.Generator</code>, <em>optional</em>) &#x2014;
Generator used when <code>sample_posterior</code> is <code>True</code>.`,name:"generator"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p><a
  href="/docs/diffusers/pr_14921/en/api/models/autoencoderkl_qwenimage21#diffusers.models.autoencoders.vae.DecoderOutput"
>DecoderOutput</a> if <code>return_dict</code> is True, otherwise a plain <code>tuple</code>.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><a
  href="/docs/diffusers/pr_14921/en/api/models/autoencoderkl_qwenimage21#diffusers.models.autoencoders.vae.DecoderOutput"
>DecoderOutput</a> or <code>tuple</code></p>
`}),n(l);var u=e(l,2),q=t(u);o(q,{name:"tiled_decode",anchor:"diffusers.AutoencoderKLQwenImage21.tiled_decode",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1454",parameters:[{name:"z",val:": Tensor"},{name:"return_dict",val:": bool = True"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.tiled_decode.z",description:"<strong>z</strong> (<code>torch.Tensor</code>) &#x2014; Input batch of latent vectors.",name:"z"},{anchor:"diffusers.AutoencoderKLQwenImage21.tiled_decode.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.vae.DecoderOutput</code> instead of a plain tuple.`,name:"return_dict"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p>If return_dict is True, a <code>~models.vae.DecoderOutput</code> is returned, otherwise a plain <code>tuple</code> is
returned.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><code>~models.vae.DecoderOutput</code> or <code>tuple</code></p>
`}),r(2),n(u);var w=e(u,2),M=t(w);o(M,{name:"tiled_encode",anchor:"diffusers.AutoencoderKLQwenImage21.tiled_encode",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1381",parameters:[{name:"x",val:": Tensor"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.tiled_encode.x",description:"<strong>x</strong> (<code>torch.Tensor</code>) &#x2014; Input batch of videos.",name:"x"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p>The latent representation of the encoded videos.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><code>torch.Tensor</code></p>
`}),r(2),n(w),n(d);var T=e(d,2);a(T,{title:"AutoencoderKLOutput",local:"diffusers.models.modeling_outputs.AutoencoderKLOutput",headingTag:"h2"});var m=e(T,2),z=t(m);o(z,{name:"class diffusers.models.modeling_outputs.AutoencoderKLOutput",anchor:"diffusers.models.modeling_outputs.AutoencoderKLOutput",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/modeling_outputs.py#L7",parameters:[{name:"latent_dist",val:": DiagonalGaussianDistribution"}],parametersDescription:[{anchor:"diffusers.models.modeling_outputs.AutoencoderKLOutput.latent_dist",description:`<strong>latent_dist</strong> (<code>DiagonalGaussianDistribution</code>) &#x2014;
Encoded outputs of <code>Encoder</code> represented as the mean and logvar of <code>DiagonalGaussianDistribution</code>.
<code>DiagonalGaussianDistribution</code> allows for sampling latents from the distribution.`,name:"latent_dist"}]}),r(2),n(m);var x=e(m,2);a(x,{title:"DecoderOutput",local:"diffusers.models.autoencoders.vae.DecoderOutput",headingTag:"h2"});var p=e(x,2),N=t(p);o(N,{name:"class diffusers.models.autoencoders.vae.DecoderOutput",anchor:"diffusers.models.autoencoders.vae.DecoderOutput",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/vae.py#L49",parameters:[{name:"sample",val:": Tensor"},{name:"commit_loss",val:": typing.Optional[torch.FloatTensor] = None"}],parametersDescription:[{anchor:"diffusers.models.autoencoders.vae.DecoderOutput.sample",description:`<strong>sample</strong> (<code>torch.Tensor</code> of shape <code>(batch_size, num_channels, height, width)</code>) &#x2014;
The decoded output sample from the last layer of the model.`,name:"sample"}]}),r(2),n(p);var W=e(p,2);G(W,{source:"https://github.com/huggingface/diffusers/blob/main/docs/source/en/api/models/autoencoderkl_qwenimage21.md"}),r(2),A(K,g),C()}export{ee as component};

Xet Storage Details

Size:: 16.1 kB
Xet hash:: 7c71fbd67f7a22c7c649c3475236efa1dd16a8200d57ed706e69451e4b6fe0c4

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.

	import"../chunks/DsnmJJEf.js";import{i as E,h as Z,C as F,H as a,a as j,D as o,E as G,s as V}from"../chunks/BLbO27j4.js";import{p as B,o as X,s as e,f as H,a as A,b as C,c as t,d as D,n as r,r as n}from"../chunks/jDjavuwI.js";const R='{"title":"AutoencoderKLQwenImage21","local":"autoencoderklqwenimage21","sections":[{"title":"AutoencoderKLQwenImage21","local":"diffusers.AutoencoderKLQwenImage21","sections":[],"depth":2},{"title":"AutoencoderKLOutput","local":"diffusers.models.modeling_outputs.AutoencoderKLOutput","sections":[],"depth":2},{"title":"DecoderOutput","local":"diffusers.models.autoencoders.vae.DecoderOutput","sections":[],"depth":2}],"depth":1}';var S=D('<meta name="hf:doc:metadata"/>'),U=D(`<p></p> <!> <!> <p>The 64-channel variational auto-encoder used by Qwen-Image 2.1. It compresses 16x spatially, and its per-channel <code>latents_mean</code> / <code>latents_std</code> are part of the config rather than a single scaling factor.</p> <!> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>A VAE model with KL loss for encoding videos into latents and decoding latent representations into videos.
	Introduced in [Qwen Image 2].</p> <p>This model inherits from <a href="/docs/diffusers/pr_14921/en/api/models/overview#diffusers.ModelMixin">ModelMixin</a>. Check the superclass documentation for it’s generic methods implemented
	for all models (such as downloading or saving).</p> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Decode a batch of images.</p></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Encode a batch of images into latents.</p></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Enable tiled VAE decoding. When this option is enabled, the VAE will split the input tensor into tiles to
	compute decoding and encoding in several steps. This is useful for saving a large amount of memory and to allow
	processing larger images.</p></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Decode a batch of images using a tiled decoder.</p></div> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Encode a batch of images using a tiled encoder.</p></div></div> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Output of AutoencoderKL encoding method.</p></div> <!> <div class="docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"><!> <p>Output of decoding method.</p></div> <!> <p></p>`,1);function ee(K,y){B(y,!1),X(()=>{new URLSearchParams(window.location.search).get("fw")}),E();var g=U();Z("nfqv89",L=>{var I=S();V(I,"content",R),A(L,I)});var f=e(H(g),2);F(f,{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"});var h=e(f,2);a(h,{title:"AutoencoderKLQwenImage21",local:"autoencoderklqwenimage21",headingTag:"h1"});var _=e(h,4);j(_,{code:"aW1wb3J0JTIwdG9yY2glMEFmcm9tJTIwZGlmZnVzZXJzJTIwaW1wb3J0JTIwQXV0b2VuY29kZXJLTFF3ZW5JbWFnZTIxJTBBJTBBdmFlJTIwJTNEJTIwQXV0b2VuY29kZXJLTFF3ZW5JbWFnZTIxLmZyb21fcHJldHJhaW5lZCglMjJRd2VuJTJGUXdlbi1JbWFnZS0yLjElMjIlMkMlMjBzdWJmb2xkZXIlM0QlMjJ2YWUlMjIlMkMlMjBkdHlwZSUzRHRvcmNoLmJmbG9hdDE2KQ==",highlighted:`<span class="hljs-keyword">import</span> torch
	<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> AutoencoderKLQwenImage21

	vae = AutoencoderKLQwenImage21.from_pretrained(<span class="hljs-string">"Qwen/Qwen-Image-2.1"</span>, subfolder=<span class="hljs-string">"vae"</span>, dtype=torch.bfloat16)`,lang:"python",wrap:!1});var v=e(_,2);a(v,{title:"AutoencoderKLQwenImage21",local:"diffusers.AutoencoderKLQwenImage21",headingTag:"h2"});var d=e(v,2),b=t(d);o(b,{name:"class diffusers.AutoencoderKLQwenImage21",anchor:"diffusers.AutoencoderKLQwenImage21",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L977",parameters:[{name:"base_dim",val:": int = 96"},{name:"decoder_base_dim",val:": int \| None = 144"},{name:"z_dim",val:": int = 64"},{name:"dim_mult",val:": list = [1, 2, 4, 8, 8]"},{name:"num_res_blocks",val:": int = 2"},{name:"attn_scales",val:": list = []"},{name:"temperal_downsample",val:": list = [False, True, True, True]"},{name:"dropout",val:": float = 0.0"},{name:"latents_mean",val:": list = [0.5126, 0.7721, -0.0631, 1.3506, -0.7855, -2.1025, -0.3458, 1.3722, 1.8873, -1.7177, -0.651, 0.2732, 0.7562, -0.6163, -1.0277, 3.8363, 2.021, 0.0472, 0.932, 2.0087, 2.4954, -0.1391, -1.4249, 1.8464, -0.5236, 1.2826, 3.7046, -1.3035, 2.7286, -1.4518, -1.9036, -1.9955, -0.0342, -1.0265, -0.7636, 3.0555, 0.0746, -3.0751, -0.1076, 1.7376, -1.0914, -1.9435, -0.2784, -1.368, 0.4809, -0.4433, 0.3764, 0.5729, -2.0595, 1.096, -1.326, -2.0211, -5.0179, 0.5275, 4.0162, 1.8505, 0.3026, 1.9373, 1.4937, 0.2632, 0.5547, -1.7121, -0.1562, 0.0304]"},{name:"latents_std",val:": list = [3.2001, 3.2936, 3.4321, 3.0091, 3.1061, 4.0379, 4.0705, 3.791, 3.0785, 3.65, 3.9308, 3.0904, 2.8778, 3.7675, 3.732, 5.0756, 3.2864, 4.0397, 3.1317, 4.0443, 2.9249, 3.9454, 3.0988, 4.2489, 3.4896, 3.8513, 3.9323, 3.4719, 3.7498, 4.283, 3.5694, 4.2467, 3.9037, 3.2947, 5.077, 3.5075, 3.27, 3.4767, 2.8063, 5.1125, 3.5327, 4.7833, 3.1286, 4.1819, 3.8527, 3.8312, 3.5605, 4.3875, 3.9624, 4.0168, 3.5643, 4.055, 5.5614, 4.2963, 4.408, 3.4959, 3.8747, 3.7608, 3.5735, 3.149, 3.7662, 3.6746, 3.4563, 3.8161]"},{name:"is_residual",val:": bool = True"},{name:"in_channels",val:": int = 4"},{name:"out_channels",val:": int = 4"},{name:"patch_size",val:": int \| None = None"},{name:"scale_factor_temporal",val:": int \| None = 8"},{name:"scale_factor_spatial",val:": int \| None = 16"}]});var s=e(b,6),Q=t(s);o(Q,{name:"decode",anchor:"diffusers.AutoencoderKLQwenImage21.decode",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1336",parameters:[{name:"z",val:": Tensor"},{name:"return_dict",val:": bool = True"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.decode.z",description:"<strong>z</strong> (<code>torch.Tensor</code>) — Input batch of latent vectors.",name:"z"},{anchor:"diffusers.AutoencoderKLQwenImage21.decode.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) —
	Whether to return a <code>~models.vae.DecoderOutput</code> instead of a plain tuple.`,name:"return_dict"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


	<p>If return_dict is True, a <code>~models.vae.DecoderOutput</code> is returned, otherwise a plain <code>tuple</code> is
	returned.</p>
	`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>


	<p><code>~models.vae.DecoderOutput</code> or <code>tuple</code></p>
	`}),r(2),n(s);var i=e(s,2),O=t(i);o(O,{name:"encode",anchor:"diffusers.AutoencoderKLQwenImage21.encode",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1276",parameters:[{name:"x",val:": Tensor"},{name:"return_dict",val:": bool = True"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.encode.x",description:"<strong>x</strong> (<code>torch.Tensor</code>) — Input batch of images.",name:"x"},{anchor:"diffusers.AutoencoderKLQwenImage21.encode.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) —
	Whether to return a <code>~models.autoencoder_kl.AutoencoderKLOutput</code> instead of a plain tuple.`,name:"return_dict"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


	<p>The latent representations of the encoded videos. If <code>return_dict</code> is True, a
	<code>~models.autoencoder_kl.AutoencoderKLOutput</code> is returned, otherwise a plain <code>tuple</code> is returned.</p>
	`}),r(2),n(i);var c=e(i,2),k=t(c);o(k,{name:"enable_tiling",anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1207",parameters:[{name:"tile_sample_min_height",val:": int \| None = None"},{name:"tile_sample_min_width",val:": int \| None = None"},{name:"tile_sample_stride_height",val:": float \| None = None"},{name:"tile_sample_stride_width",val:": float \| None = None"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling.tile_sample_min_height",description:`<strong>tile_sample_min_height</strong> (<code>int</code>, <em>optional</em>) —
	The minimum height required for a sample to be separated into tiles across the height dimension.`,name:"tile_sample_min_height"},{anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling.tile_sample_min_width",description:`<strong>tile_sample_min_width</strong> (<code>int</code>, <em>optional</em>) —
	The minimum width required for a sample to be separated into tiles across the width dimension.`,name:"tile_sample_min_width"},{anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling.tile_sample_stride_height",description:`<strong>tile_sample_stride_height</strong> (<code>int</code>, <em>optional</em>) —
	The minimum amount of overlap between two consecutive vertical tiles. This is to ensure that there are
	no tiling artifacts produced across the height dimension.`,name:"tile_sample_stride_height"},{anchor:"diffusers.AutoencoderKLQwenImage21.enable_tiling.tile_sample_stride_width",description:`<strong>tile_sample_stride_width</strong> (<code>int</code>, <em>optional</em>) —
	The stride between two consecutive horizontal tiles. This is to ensure that there are no tiling
	artifacts produced across the width dimension.`,name:"tile_sample_stride_width"}]}),r(2),n(c);var l=e(c,2),J=t(l);o(J,{name:"forward",anchor:"diffusers.AutoencoderKLQwenImage21.forward",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1532",parameters:[{name:"sample",val:": Tensor"},{name:"sample_posterior",val:": bool = False"},{name:"return_dict",val:": bool = True"},{name:"generator",val:": typing.Optional[torch.Generator] = None"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.forward.sample",description:"<strong>sample</strong> (<code>torch.Tensor</code>) — Input sample.",name:"sample"},{anchor:"diffusers.AutoencoderKLQwenImage21.forward.sample_posterior",description:`<strong>sample_posterior</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>False</code>) —
	Whether to sample from the posterior instead of taking its mode.`,name:"sample_posterior"},{anchor:"diffusers.AutoencoderKLQwenImage21.forward.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) —
	Whether or not to return a <code>DecoderOutput</code> instead of a plain tuple.`,name:"return_dict"},{anchor:"diffusers.AutoencoderKLQwenImage21.forward.generator",description:`<strong>generator</strong> (<code>torch.Generator</code>, <em>optional</em>) —
	Generator used when <code>sample_posterior</code> is <code>True</code>.`,name:"generator"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


	<p><a
	href="/docs/diffusers/pr_14921/en/api/models/autoencoderkl_qwenimage21#diffusers.models.autoencoders.vae.DecoderOutput"
	>DecoderOutput</a> if <code>return_dict</code> is True, otherwise a plain <code>tuple</code>.</p>
	`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>


	<p><a
	href="/docs/diffusers/pr_14921/en/api/models/autoencoderkl_qwenimage21#diffusers.models.autoencoders.vae.DecoderOutput"
	>DecoderOutput</a> or <code>tuple</code></p>
	`}),n(l);var u=e(l,2),q=t(u);o(q,{name:"tiled_decode",anchor:"diffusers.AutoencoderKLQwenImage21.tiled_decode",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1454",parameters:[{name:"z",val:": Tensor"},{name:"return_dict",val:": bool = True"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.tiled_decode.z",description:"<strong>z</strong> (<code>torch.Tensor</code>) — Input batch of latent vectors.",name:"z"},{anchor:"diffusers.AutoencoderKLQwenImage21.tiled_decode.return_dict",description:`<strong>return_dict</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) —
	Whether or not to return a <code>~models.vae.DecoderOutput</code> instead of a plain tuple.`,name:"return_dict"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


	<p>If return_dict is True, a <code>~models.vae.DecoderOutput</code> is returned, otherwise a plain <code>tuple</code> is
	returned.</p>
	`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>


	<p><code>~models.vae.DecoderOutput</code> or <code>tuple</code></p>
	`}),r(2),n(u);var w=e(u,2),M=t(w);o(M,{name:"tiled_encode",anchor:"diffusers.AutoencoderKLQwenImage21.tiled_encode",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/autoencoder_kl_qwenimage21.py#L1381",parameters:[{name:"x",val:": Tensor"}],parametersDescription:[{anchor:"diffusers.AutoencoderKLQwenImage21.tiled_encode.x",description:"<strong>x</strong> (<code>torch.Tensor</code>) — Input batch of videos.",name:"x"}],returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>


	<p>The latent representation of the encoded videos.</p>
	`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>


	<p><code>torch.Tensor</code></p>
	`}),r(2),n(w),n(d);var T=e(d,2);a(T,{title:"AutoencoderKLOutput",local:"diffusers.models.modeling_outputs.AutoencoderKLOutput",headingTag:"h2"});var m=e(T,2),z=t(m);o(z,{name:"class diffusers.models.modeling_outputs.AutoencoderKLOutput",anchor:"diffusers.models.modeling_outputs.AutoencoderKLOutput",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/modeling_outputs.py#L7",parameters:[{name:"latent_dist",val:": DiagonalGaussianDistribution"}],parametersDescription:[{anchor:"diffusers.models.modeling_outputs.AutoencoderKLOutput.latent_dist",description:`<strong>latent_dist</strong> (<code>DiagonalGaussianDistribution</code>) —
	Encoded outputs of <code>Encoder</code> represented as the mean and logvar of <code>DiagonalGaussianDistribution</code>.
	<code>DiagonalGaussianDistribution</code> allows for sampling latents from the distribution.`,name:"latent_dist"}]}),r(2),n(m);var x=e(m,2);a(x,{title:"DecoderOutput",local:"diffusers.models.autoencoders.vae.DecoderOutput",headingTag:"h2"});var p=e(x,2),N=t(p);o(N,{name:"class diffusers.models.autoencoders.vae.DecoderOutput",anchor:"diffusers.models.autoencoders.vae.DecoderOutput",source:"https://github.com/huggingface/diffusers/blob/vr_14921/src/diffusers/models/autoencoders/vae.py#L49",parameters:[{name:"sample",val:": Tensor"},{name:"commit_loss",val:": typing.Optional[torch.FloatTensor] = None"}],parametersDescription:[{anchor:"diffusers.models.autoencoders.vae.DecoderOutput.sample",description:`<strong>sample</strong> (<code>torch.Tensor</code> of shape <code>(batch_size, num_channels, height, width)</code>) —
	The decoded output sample from the last layer of the model.`,name:"sample"}]}),r(2),n(p);var W=e(p,2);G(W,{source:"https://github.com/huggingface/diffusers/blob/main/docs/source/en/api/models/autoencoderkl_qwenimage21.md"}),r(2),A(K,g),C()}export{ee as component};