Buckets:

HuggingFaceDocBuilder's picture
download
raw
80.3 kB
import{s as Ct,o as wt,n as ee}from"../chunks/scheduler.31fdf58d.js";import{S as Ut,i as bt,e as m,s as a,c as g,h as vt,a as h,d as r,b as i,f as F,j as C,g as f,k,l,m as p,n as _,t as y,o as M,p as T}from"../chunks/index.2f76fdf0.js";import{T as Mt}from"../chunks/Tip.8d349121.js";import{C as Jt}from"../chunks/CopyLLMTxtMenu.53b607bf.js";import{D as A}from"../chunks/Docstring.7acc6835.js";import{C as Le}from"../chunks/CodeBlock.e52df5d6.js";import{E as _o}from"../chunks/ExampleCodeBlock.f9704f52.js";import{H as K,E as It}from"../chunks/MermaidChart.svelte_svelte_type_style_lang.08750ec0.js";import{H as jt,a as Tt}from"../chunks/HfOption.fb051768.js";function $t(U){let o,u;return o=new Le({props:{code:"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",highlighted:`
<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoModelForImageTextToText, AutoProcessor
model_id = <span class="hljs-string">&quot;CohereLabs/command-a-vision-07-2025&quot;</span>
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForImageTextToText.from_pretrained(
model_id, device_map=<span class="hljs-string">&quot;auto&quot;</span>
)
<span class="hljs-comment"># Format message with the Command-A-Vision chat template</span>
messages = [
{
<span class="hljs-string">&quot;role&quot;</span>: <span class="hljs-string">&quot;user&quot;</span>,
<span class="hljs-string">&quot;content&quot;</span>: [
{
<span class="hljs-string">&quot;type&quot;</span>: <span class="hljs-string">&quot;image&quot;</span>,
<span class="hljs-string">&quot;url&quot;</span>: <span class="hljs-string">&quot;https://images.pexels.com/photos/1108099/pexels-photo-1108099.jpeg&quot;</span>,
},
{<span class="hljs-string">&quot;type&quot;</span>: <span class="hljs-string">&quot;text&quot;</span>, <span class="hljs-string">&quot;text&quot;</span>: <span class="hljs-string">&quot;what is in this image?&quot;</span>},
],
},
]
inputs = processor.apply_chat_template(
messages,
padding=<span class="hljs-literal">True</span>,
add_generation_prompt=<span class="hljs-literal">True</span>,
tokenize=<span class="hljs-literal">True</span>,
return_dict=<span class="hljs-literal">True</span>,
return_tensors=<span class="hljs-string">&quot;pt&quot;</span>,
).to(model.device)
gen_tokens = model.generate(
**inputs,
max_new_tokens=<span class="hljs-number">300</span>,
do_sample=<span class="hljs-literal">True</span>,
temperature=<span class="hljs-number">0.3</span>,
)
<span class="hljs-built_in">print</span>(
processor.tokenizer.decode(
gen_tokens[<span class="hljs-number">0</span>][inputs.input_ids.shape[<span class="hljs-number">1</span>] :], skip_special_tokens=<span class="hljs-literal">True</span>
)
)`,lang:"python",wrap:!1}}),{c(){g(o.$$.fragment)},l(s){f(o.$$.fragment,s)},m(s,c){_(o,s,c),u=!0},p:ee,i(s){u||(y(o.$$.fragment,s),u=!0)},o(s){M(o.$$.fragment,s),u=!1},d(s){T(o,s)}}}function kt(U){let o,u;return o=new Le({props:{code:"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",highlighted:`<span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> pipeline
pipe = pipeline(model=<span class="hljs-string">&quot;CohereLabs/command-a-vision-07-2025&quot;</span>, task=<span class="hljs-string">&quot;image-text-to-text&quot;</span>, device_map=<span class="hljs-string">&quot;auto&quot;</span>)
messages = [
{
<span class="hljs-string">&quot;role&quot;</span>: <span class="hljs-string">&quot;user&quot;</span>,
<span class="hljs-string">&quot;content&quot;</span>: [
{
<span class="hljs-string">&quot;type&quot;</span>: <span class="hljs-string">&quot;image&quot;</span>,
<span class="hljs-string">&quot;url&quot;</span>: <span class="hljs-string">&quot;https://media.istockphoto.com/id/458012057/photo/istanbul-turkey.jpg?s=612x612&amp;w=0&amp;k=20&amp;c=qogAOVvkpfUyqLUMr_XJQyq-HkACXyYUSZbKhBlPrxo=&quot;</span>,
},
{<span class="hljs-string">&quot;type&quot;</span>: <span class="hljs-string">&quot;text&quot;</span>, <span class="hljs-string">&quot;text&quot;</span>: <span class="hljs-string">&quot;Where was this taken ?&quot;</span>},
],
},
]
outputs = pipe(text=messages, max_new_tokens=<span class="hljs-number">300</span>, return_full_text=<span class="hljs-literal">False</span>)
<span class="hljs-built_in">print</span>(outputs)`,lang:"python",wrap:!1}}),{c(){g(o.$$.fragment)},l(s){f(o.$$.fragment,s)},m(s,c){_(o,s,c),u=!0},p:ee,i(s){u||(y(o.$$.fragment,s),u=!0)},o(s){M(o.$$.fragment,s),u=!1},d(s){T(o,s)}}}function Vt(U){let o,u,s,c;return o=new Tt({props:{id:"usage",option:"AutoModel",$$slots:{default:[$t]},$$scope:{ctx:U}}}),s=new Tt({props:{id:"usage",option:"Pipeline",$$slots:{default:[kt]},$$scope:{ctx:U}}}),{c(){g(o.$$.fragment),u=a(),g(s.$$.fragment)},l(d){f(o.$$.fragment,d),u=i(d),f(s.$$.fragment,d)},m(d,t){_(o,d,t),p(d,u,t),_(s,d,t),c=!0},p(d,t){const w={};t&2&&(w.$$scope={dirty:t,ctx:d}),o.$set(w);const N={};t&2&&(N.$$scope={dirty:t,ctx:d}),s.$set(N)},i(d){c||(y(o.$$.fragment,d),y(s.$$.fragment,d),c=!0)},o(d){M(o.$$.fragment,d),M(s.$$.fragment,d),c=!1},d(d){d&&r(u),T(o,d),T(s,d)}}}function xt(U){let o,u=`Although the recipe for forward pass needs to be defined within this function, one should call the <code>Module</code>
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.`;return{c(){o=m("p"),o.innerHTML=u},l(s){o=h(s,"P",{"data-svelte-h":!0}),C(o)!=="svelte-fincs2"&&(o.innerHTML=u)},m(s,c){p(s,o,c)},p:ee,d(s){s&&r(o)}}}function zt(U){let o,u="Example:",s,c,d;return c=new Le({props:{code:"ZnJvbSUyMHRyYW5zZm9ybWVycyUyMGltcG9ydCUyMEF1dG9Qcm9jZXNzb3IlMkMlMjBDb2hlcmUyVmlzaW9uRm9yQ29uZGl0aW9uYWxHZW5lcmF0aW9uJTBBaW1wb3J0JTIwdG9yY2glMEElMEFwcm9jZXNzb3IlMjAlM0QlMjBBdXRvUHJvY2Vzc29yLmZyb21fcHJldHJhaW5lZCglMjJDb2hlcmVMYWJzJTJGY29tbWFuZC1hLXZpc2lvbi0wNy0yMDI1JTIyJTJDJTIwdXNlX2Zhc3QlM0RUcnVlKSUwQW1vZGVsJTIwJTNEJTIwQ29oZXJlMlZpc2lvbkZvckNvbmRpdGlvbmFsR2VuZXJhdGlvbi5mcm9tX3ByZXRyYWluZWQoJTIyQ29oZXJlTGFicyUyRmNvbW1hbmQtYS12aXNpb24tMDctMjAyNSUyMiUyQyUyMGRldmljZV9tYXAlM0QlMjJhdXRvJTIyKSUwQSUwQW1lc3NhZ2VzJTIwJTNEJTIwJTVCJTBBJTIwJTIwJTIwJTIwJTdCJTBBJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIycm9sZSUyMiUzQSUyMCUyMnVzZXIlMjIlMkMlMEElMjAlMjAlMjAlMjAlMjAlMjAlMjAlMjAlMjJjb250ZW50JTIyJTNBJTIwJTVCJTBBJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTdCJTBBJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIydHlwZSUyMiUzQSUyMCUyMmltYWdlJTIyJTJDJTBBJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIydXJsJTIyJTNBJTIwJTIyaHR0cHMlM0ElMkYlMkZpbWFnZXMucGV4ZWxzLmNvbSUyRnBob3RvcyUyRjExMDgwOTklMkZwZXhlbHMtcGhvdG8tMTEwODA5OS5qcGVnJTIyJTJDJTBBJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTdEJTJDJTBBJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTIwJTdCJTIydHlwZSUyMiUzQSUyMCUyMnRleHQlMjIlMkMlMjAlMjJ0ZXh0JTIyJTNBJTIwJTIyd2hhdCUyMGlzJTIwaW4lMjB0aGlzJTIwaW1hZ2UlM0YlMjIlN0QlMkMlMEElMjAlMjAlMjAlMjAlMjAlMjAlMjAlMjAlNUQlMkMlMEElMjAlMjAlMjAlMjAlN0QlMkMlMEElNUQlMEElMEFpbnB1dHMlMjAlM0QlMjBwcm9jZXNzb3IuYXBwbHlfY2hhdF90ZW1wbGF0ZSglMEElMjAlMjAlMjAlMjBtZXNzYWdlcyUyQyUyMHBhZGRpbmclM0RUcnVlJTJDJTIwYWRkX2dlbmVyYXRpb25fcHJvbXB0JTNEVHJ1ZSUyQyUyMHRva2VuaXplJTNEVHJ1ZSUyQyUyMHJldHVybl9kaWN0JTNEVHJ1ZSUyQyUyMHJldHVybl90ZW5zb3JzJTNEJTIycHQlMjIlMkMlMEEpLnRvKG1vZGVsLmRldmljZSklMEElMEFnZW5fdG9rZW5zJTIwJTNEJTIwbW9kZWwuZ2VuZXJhdGUoKippbnB1dHMlMkMlMjBtYXhfbmV3X3Rva2VucyUzRDMwMCUyQyUyMGRvX3NhbXBsZSUzRFRydWUlMkMlMjB0ZW1wZXJhdHVyZSUzRDAuMyklMEFwcm9jZXNzb3IudG9rZW5pemVyLmRlY29kZShnZW5fdG9rZW5zJTVCMCU1RCU1QmlucHV0cy5pbnB1dF9pZHMuc2hhcGUlNUIxJTVEJTNBJTVEJTJDJTIwc2tpcF9zcGVjaWFsX3Rva2VucyUzRFRydWUp",highlighted:`<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoProcessor, Cohere2VisionForConditionalGeneration
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">import</span> torch
<span class="hljs-meta">&gt;&gt;&gt; </span>processor = AutoProcessor.from_pretrained(<span class="hljs-string">&quot;CohereLabs/command-a-vision-07-2025&quot;</span>, use_fast=<span class="hljs-literal">True</span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>model = Cohere2VisionForConditionalGeneration.from_pretrained(<span class="hljs-string">&quot;CohereLabs/command-a-vision-07-2025&quot;</span>, device_map=<span class="hljs-string">&quot;auto&quot;</span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>messages = [
<span class="hljs-meta">... </span> {
<span class="hljs-meta">... </span> <span class="hljs-string">&quot;role&quot;</span>: <span class="hljs-string">&quot;user&quot;</span>,
<span class="hljs-meta">... </span> <span class="hljs-string">&quot;content&quot;</span>: [
<span class="hljs-meta">... </span> {
<span class="hljs-meta">... </span> <span class="hljs-string">&quot;type&quot;</span>: <span class="hljs-string">&quot;image&quot;</span>,
<span class="hljs-meta">... </span> <span class="hljs-string">&quot;url&quot;</span>: <span class="hljs-string">&quot;https://images.pexels.com/photos/1108099/pexels-photo-1108099.jpeg&quot;</span>,
<span class="hljs-meta">... </span> },
<span class="hljs-meta">... </span> {<span class="hljs-string">&quot;type&quot;</span>: <span class="hljs-string">&quot;text&quot;</span>, <span class="hljs-string">&quot;text&quot;</span>: <span class="hljs-string">&quot;what is in this image?&quot;</span>},
<span class="hljs-meta">... </span> ],
<span class="hljs-meta">... </span> },
<span class="hljs-meta">... </span>]
<span class="hljs-meta">&gt;&gt;&gt; </span>inputs = processor.apply_chat_template(
<span class="hljs-meta">... </span> messages, padding=<span class="hljs-literal">True</span>, add_generation_prompt=<span class="hljs-literal">True</span>, tokenize=<span class="hljs-literal">True</span>, return_dict=<span class="hljs-literal">True</span>, return_tensors=<span class="hljs-string">&quot;pt&quot;</span>,
<span class="hljs-meta">... </span>).to(model.device)
<span class="hljs-meta">&gt;&gt;&gt; </span>gen_tokens = model.generate(**inputs, max_new_tokens=<span class="hljs-number">300</span>, do_sample=<span class="hljs-literal">True</span>, temperature=<span class="hljs-number">0.3</span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>processor.tokenizer.decode(gen_tokens[<span class="hljs-number">0</span>][inputs.input_ids.shape[<span class="hljs-number">1</span>]:], skip_special_tokens=<span class="hljs-literal">True</span>)`,lang:"python",wrap:!1}}),{c(){o=m("p"),o.textContent=u,s=a(),g(c.$$.fragment)},l(t){o=h(t,"P",{"data-svelte-h":!0}),C(o)!=="svelte-11lpom8"&&(o.textContent=u),s=i(t),f(c.$$.fragment,t)},m(t,w){p(t,o,w),p(t,s,w),_(c,t,w),d=!0},p:ee,i(t){d||(y(c.$$.fragment,t),d=!0)},o(t){M(c.$$.fragment,t),d=!1},d(t){t&&(r(o),r(s)),T(c,t)}}}function Ft(U){let o,u="Example:",s,c,d;return c=new Le({props:{code:"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",highlighted:`<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> PIL <span class="hljs-keyword">import</span> Image
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoProcessor, Cohere2VisionForConditionalGeneration
<span class="hljs-meta">&gt;&gt;&gt; </span>model = Cohere2VisionForConditionalGeneration.from_pretrained(<span class="hljs-string">&quot;CohereLabs/command-a-vision-07-2025&quot;</span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>processor = AutoProcessor.from_pretrained(<span class="hljs-string">&quot;CohereLabs/command-a-vision-07-2025&quot;</span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>messages = [
<span class="hljs-meta">... </span> {
<span class="hljs-meta">... </span> <span class="hljs-string">&quot;role&quot;</span>: <span class="hljs-string">&quot;user&quot;</span>, <span class="hljs-string">&quot;content&quot;</span>: [
<span class="hljs-meta">... </span> {<span class="hljs-string">&quot;type&quot;</span>: <span class="hljs-string">&quot;image&quot;</span>, <span class="hljs-string">&quot;url&quot;</span>: <span class="hljs-string">&quot;https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg&quot;</span>},
<span class="hljs-meta">... </span> {<span class="hljs-string">&quot;type&quot;</span>: <span class="hljs-string">&quot;text&quot;</span>, <span class="hljs-string">&quot;text&quot;</span>: <span class="hljs-string">&quot;Where is the cat standing?&quot;</span>},
<span class="hljs-meta">... </span> ]
<span class="hljs-meta">... </span> },
<span class="hljs-meta">... </span>]
<span class="hljs-meta">&gt;&gt;&gt; </span>inputs = processor.apply_chat_template(
<span class="hljs-meta">... </span> messages,
<span class="hljs-meta">... </span> tokenize=<span class="hljs-literal">True</span>,
<span class="hljs-meta">... </span> return_dict=<span class="hljs-literal">True</span>,
<span class="hljs-meta">... </span> return_tensors=<span class="hljs-string">&quot;pt&quot;</span>,
<span class="hljs-meta">... </span> add_generation_prompt=<span class="hljs-literal">True</span>
<span class="hljs-meta">... </span>)
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-comment"># Generate</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>generate_ids = model.generate(**inputs)
<span class="hljs-meta">&gt;&gt;&gt; </span>processor.batch_decode(generate_ids, skip_special_tokens=<span class="hljs-literal">True</span>)[<span class="hljs-number">0</span>]`,lang:"python",wrap:!1}}),{c(){o=m("p"),o.textContent=u,s=a(),g(c.$$.fragment)},l(t){o=h(t,"P",{"data-svelte-h":!0}),C(o)!=="svelte-11lpom8"&&(o.textContent=u),s=i(t),f(c.$$.fragment,t)},m(t,w){p(t,o,w),p(t,s,w),_(c,t,w),d=!0},p:ee,i(t){d||(y(c.$$.fragment,t),d=!0)},o(t){M(c.$$.fragment,t),d=!1},d(t){t&&(r(o),r(s)),T(c,t)}}}function Bt(U){let o,u=`Although the recipe for forward pass needs to be defined within this function, one should call the <code>Module</code>
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.`;return{c(){o=m("p"),o.innerHTML=u},l(s){o=h(s,"P",{"data-svelte-h":!0}),C(o)!=="svelte-fincs2"&&(o.innerHTML=u)},m(s,c){p(s,o,c)},p:ee,d(s){s&&r(o)}}}function qt(U){let o,u="Example:",s,c,d;return c=new Le({props:{code:"",highlighted:"",lang:"python",wrap:!1}}),{c(){o=m("p"),o.textContent=u,s=a(),g(c.$$.fragment)},l(t){o=h(t,"P",{"data-svelte-h":!0}),C(o)!=="svelte-11lpom8"&&(o.textContent=u),s=i(t),f(c.$$.fragment,t)},m(t,w){p(t,o,w),p(t,s,w),_(c,t,w),d=!0},p:ee,i(t){d||(y(c.$$.fragment,t),d=!0)},o(t){M(c.$$.fragment,t),d=!1},d(t){t&&(r(o),r(s)),T(c,t)}}}function Pt(U){let o,u="Example:",s,c,d;return c=new Le({props:{code:"",highlighted:"",lang:"python",wrap:!1}}),{c(){o=m("p"),o.textContent=u,s=a(),g(c.$$.fragment)},l(t){o=h(t,"P",{"data-svelte-h":!0}),C(o)!=="svelte-11lpom8"&&(o.textContent=u),s=i(t),f(c.$$.fragment,t)},m(t,w){p(t,o,w),p(t,s,w),_(c,t,w),d=!0},p:ee,i(t){d||(y(c.$$.fragment,t),d=!0)},o(t){M(c.$$.fragment,t),d=!1},d(t){t&&(r(o),r(s)),T(c,t)}}}function Zt(U){let o,u,s,c,d,t="<em>This model was contributed to Hugging Face Transformers on 2025-07-31.</em>",w,N,Xe,oe,Se,R,Xo='<img alt="FlashAttention" src="https://img.shields.io/badge/%E2%9A%A1%EF%B8%8E%20FlashAttention-eae0c8?style=flat"/> <img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&amp;logo=pytorch&amp;logoColor=white"/> <img alt="Tensor parallelism" src="https://img.shields.io/badge/Tensor%20parallelism-06b6d4?style=flat&amp;logoColor=white"/>',De,te,Ye,se,So='Command A Vision (<a href="https://cohere.com/blog/command-a-vision" rel="nofollow">blog post</a>) is a state-of-the-art multimodal model designed to seamlessly integrate visual and textual information for a wide range of applications. By combining advanced computer vision techniques with natural language processing capabilities, Command A Vision enables users to analyze, understand, and generate insights from both visual and textual data.',Oe,ne,Do="The model excels at tasks including image captioning, visual question answering, document understanding, and chart understanding. This makes it a versatile tool for AI practitioners. Its ability to process complex visual and textual inputs makes it useful in settings where text-only representations are imprecise or unavailable, like real-world image understanding and graphics-heavy document processing.",Ke,re,Yo="Command A Vision is built upon a robust architecture that leverages the latest advancements in VLMs. It’s highly performant and efficient, even when dealing with large-scale datasets. The model’s flexibility makes it suitable for a wide range of use cases, from content moderation and image search to medical imaging analysis and robotics.",eo,ae,oo,ie,Oo="The model and image processor can be loaded as follows:",to,Q,so,le,no,q,ce,yo,ve,Ko=`This is the configuration class to store the configuration of a Cohere2VisionModel. It is used to instantiate a Cohere2 Vision
model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
defaults will yield a similar configuration to that of the <a href="https://huggingface.co/CohereLabs/command-a-vision-07-2025" rel="nofollow">CohereLabs/command-a-vision-07-2025</a>`,Mo,Je,et=`Configuration objects inherit from <a href="/docs/transformers/pr_43265/en/main_classes/configuration#transformers.PreTrainedConfig">PreTrainedConfig</a> and can be used to control the model outputs. Read the
documentation from <a href="/docs/transformers/pr_43265/en/main_classes/configuration#transformers.PreTrainedConfig">PreTrainedConfig</a> for more information.`,ro,de,ao,b,pe,To,Ie,ot="The COHERE2_VISION model which consists of a vision backbone and a language model.",Co,je,tt=`This model inherits from <a href="/docs/transformers/pr_43265/en/main_classes/model#transformers.PreTrainedModel">PreTrainedModel</a>. Check the superclass documentation for the generic methods the
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
etc.)`,wo,$e,st=`This model is also a PyTorch <a href="https://pytorch.org/docs/stable/nn.html#torch.nn.Module" rel="nofollow">torch.nn.Module</a> subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
and behavior.`,Uo,j,me,bo,ke,nt='The <a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionForConditionalGeneration">Cohere2VisionForConditionalGeneration</a> forward method, overrides the <code>__call__</code> special method.',vo,L,Jo,Ve,rt=`<li><p><strong>loss</strong> (<code>torch.FloatTensor</code> of shape <code>(1,)</code>, <em>optional</em>, returned when <code>labels</code> is provided) — Language modeling loss (for next-token prediction).</p></li> <li><p><strong>logits</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, sequence_length, config.vocab_size)</code>) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).</p></li> <li><p><strong>past_key_values</strong> (<code>Cache</code>, <em>optional</em>, returned when <code>use_cache=True</code> is passed or when <code>config.use_cache=True</code>) — It is a <a href="/docs/transformers/pr_43265/en/internal/generation_utils#transformers.Cache">Cache</a> instance. For more details, see our <a href="https://huggingface.co/docs/transformers/en/kv_cache" rel="nofollow">kv cache guide</a>.</p> <p>Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see
<code>past_key_values</code> input) to speed up sequential decoding.</p></li> <li><p><strong>hidden_states</strong> (<code>tuple[torch.FloatTensor]</code>, <em>optional</em>, returned when <code>output_hidden_states=True</code> is passed or when <code>config.output_hidden_states=True</code>) — Tuple of <code>torch.FloatTensor</code> (one for the output of the embeddings, if the model has an embedding layer, +
one for the output of each layer) of shape <code>(batch_size, sequence_length, hidden_size)</code>.</p> <p>Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.</p></li> <li><p><strong>attentions</strong> (<code>tuple[torch.FloatTensor]</code>, <em>optional</em>, returned when <code>output_attentions=True</code> is passed or when <code>config.output_attentions=True</code>) — Tuple of <code>torch.FloatTensor</code> (one for each layer) of shape <code>(batch_size, num_heads, sequence_length, sequence_length)</code>.</p> <p>Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
heads.</p></li> <li><p><strong>image_hidden_states</strong> (<code>torch.FloatTensor</code>, <em>optional</em>) — A <code>torch.FloatTensor</code> of size <code>(batch_size, num_images, sequence_length, hidden_size)</code>.
image_hidden_states of the model produced by the vision encoder and after projecting the last hidden state.</p></li>`,Io,H,jo,G,he,$o,xe,at=`<li><p><strong>last_hidden_state</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, sequence_length, hidden_size)</code>) — Sequence of hidden-states at the output of the last layer of the model.</p></li> <li><p><strong>pooler_output</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, hidden_size)</code>) — Last layer hidden-state of the first token of the sequence (classification token) after further processing
through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns
the classification token after processing through a linear layer and a tanh activation function. The linear
layer weights are trained from the next sentence prediction (classification) objective during pretraining.</p></li> <li><p><strong>hidden_states</strong> (<code>tuple(torch.FloatTensor)</code>, <em>optional</em>, returned when <code>output_hidden_states=True</code> is passed or when <code>config.output_hidden_states=True</code>) — Tuple of <code>torch.FloatTensor</code> (one for the output of the embeddings, if the model has an embedding layer, +
one for the output of each layer) of shape <code>(batch_size, sequence_length, hidden_size)</code>.</p> <p>Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.</p></li> <li><p><strong>attentions</strong> (<code>tuple(torch.FloatTensor)</code>, <em>optional</em>, returned when <code>output_attentions=True</code> is passed or when <code>config.output_attentions=True</code>) — Tuple of <code>torch.FloatTensor</code> (one for each layer) of shape <code>(batch_size, num_heads, sequence_length, sequence_length)</code>.</p> <p>Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
heads.</p></li>`,ko,X,io,ue,lo,v,ge,Vo,ze,it="The Cohere2Vision model which consists of a vision backbone and a language model, without a language modeling head.",xo,Fe,lt=`This model inherits from <a href="/docs/transformers/pr_43265/en/main_classes/model#transformers.PreTrainedModel">PreTrainedModel</a>. Check the superclass documentation for the generic methods the
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
etc.)`,zo,Be,ct=`This model is also a PyTorch <a href="https://pytorch.org/docs/stable/nn.html#torch.nn.Module" rel="nofollow">torch.nn.Module</a> subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
and behavior.`,Fo,$,fe,Bo,qe,dt='The <a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionModel">Cohere2VisionModel</a> forward method, overrides the <code>__call__</code> special method.',qo,S,Po,Pe,pt=`<li><p><strong>last_hidden_state</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, sequence_length, hidden_size)</code>) — Sequence of hidden-states at the output of the last layer of the model.</p> <p>If <code>past_key_values</code> is used only the last hidden-state of the sequences of shape <code>(batch_size, 1, hidden_size)</code> is output.</p></li> <li><p><strong>past_key_values</strong> (<code>Cache</code>, <em>optional</em>, returned when <code>use_cache=True</code> is passed or when <code>config.use_cache=True</code>) — It is a <a href="/docs/transformers/pr_43265/en/internal/generation_utils#transformers.Cache">Cache</a> instance. For more details, see our <a href="https://huggingface.co/docs/transformers/en/kv_cache" rel="nofollow">kv cache guide</a>.</p> <p>Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see
<code>past_key_values</code> input) to speed up sequential decoding.</p></li> <li><p><strong>hidden_states</strong> (<code>tuple(torch.FloatTensor)</code>, <em>optional</em>, returned when <code>output_hidden_states=True</code> is passed or when <code>config.output_hidden_states=True</code>) — Tuple of <code>torch.FloatTensor</code> (one for the output of the embeddings, if the model has an embedding layer, +
one for the output of each layer) of shape <code>(batch_size, sequence_length, hidden_size)</code>.</p> <p>Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.</p></li> <li><p><strong>attentions</strong> (<code>tuple(torch.FloatTensor)</code>, <em>optional</em>, returned when <code>output_attentions=True</code> is passed or when <code>config.output_attentions=True</code>) — Tuple of <code>torch.FloatTensor</code> (one for each layer) of shape <code>(batch_size, num_heads, sequence_length, sequence_length)</code>.</p> <p>Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
heads.</p></li> <li><p><strong>image_hidden_states</strong> (<code>torch.FloatTensor</code>, <em>optional</em>) — A <code>torch.FloatTensor</code> of size <code>(batch_size, num_images, sequence_length, hidden_size)</code>.
image_hidden_states of the model produced by the vision encoder and after projecting the last hidden state.</p></li>`,Zo,D,Ao,B,_e,No,Ze,mt="Obtains image last hidden states from the vision tower and apply multimodal projection.",Go,Ae,ht=`<li><p><strong>last_hidden_state</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, sequence_length, hidden_size)</code>) — Sequence of hidden-states at the output of the last layer of the model.</p></li> <li><p><strong>pooler_output</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, hidden_size)</code>) — Last layer hidden-state of the first token of the sequence (classification token) after further processing
through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns
the classification token after processing through a linear layer and a tanh activation function. The linear
layer weights are trained from the next sentence prediction (classification) objective during pretraining.</p></li> <li><p><strong>hidden_states</strong> (<code>tuple(torch.FloatTensor)</code>, <em>optional</em>, returned when <code>output_hidden_states=True</code> is passed or when <code>config.output_hidden_states=True</code>) — Tuple of <code>torch.FloatTensor</code> (one for the output of the embeddings, if the model has an embedding layer, +
one for the output of each layer) of shape <code>(batch_size, sequence_length, hidden_size)</code>.</p> <p>Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.</p></li> <li><p><strong>attentions</strong> (<code>tuple(torch.FloatTensor)</code>, <em>optional</em>, returned when <code>output_attentions=True</code> is passed or when <code>config.output_attentions=True</code>) — Tuple of <code>torch.FloatTensor</code> (one for each layer) of shape <code>(batch_size, num_heads, sequence_length, sequence_length)</code>.</p> <p>Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
heads.</p></li>`,Eo,Y,co,ye,po,P,Me,Wo,Ne,ut="Constructs a Cohere2VisionImageProcessor image processor.",Ro,Ge,Te,mo,Ce,ho,V,we,Qo,Ee,gt="Constructs a Cohere2VisionProcessor which wraps a image processor and a tokenizer into a single processor.",Lo,We,ft=`<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionProcessor">Cohere2VisionProcessor</a> offers all the functionalities of <a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">Cohere2VisionImageProcessor</a> and <code>tokenizer_class</code>. See the
<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">~Cohere2VisionImageProcessor</a> and <code>~tokenizer_class</code> for more information.`,Ho,Re,Ue,uo,be,go,He,fo;return N=new Jt({props:{containerStyle:"float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"}}),oe=new K({props:{title:"Command A Vision",local:"command-a-vision",headingTag:"h1"}}),te=new K({props:{title:"Overview",local:"overview",headingTag:"h2"}}),ae=new K({props:{title:"Usage tips",local:"usage-tips",headingTag:"h2"}}),Q=new jt({props:{id:"usage",options:["AutoModel","Pipeline"],$$slots:{default:[Vt]},$$scope:{ctx:U}}}),le=new K({props:{title:"Cohere2VisionConfig",local:"transformers.Cohere2VisionConfig",headingTag:"h2"}}),ce=new A({props:{name:"class transformers.Cohere2VisionConfig",anchor:"transformers.Cohere2VisionConfig",parameters:[{name:"transformers_version",val:": str | None = None"},{name:"architectures",val:": list[str] | None = None"},{name:"output_hidden_states",val:": bool | None = False"},{name:"return_dict",val:": bool | None = True"},{name:"dtype",val:": typing.Union[str, ForwardRef('torch.dtype'), NoneType] = None"},{name:"chunk_size_feed_forward",val:": int = 0"},{name:"is_encoder_decoder",val:": bool = False"},{name:"id2label",val:": dict[int, str] | dict[str, str] | None = None"},{name:"label2id",val:": dict[str, int] | dict[str, str] | None = None"},{name:"problem_type",val:": typing.Optional[typing.Literal['regression', 'single_label_classification', 'multi_label_classification']] = None"},{name:"vision_config",val:": dict | transformers.configuration_utils.PreTrainedConfig | None = None"},{name:"text_config",val:": dict | transformers.configuration_utils.PreTrainedConfig | None = None"},{name:"downsample_factor",val:": int = 2"},{name:"image_token_id",val:": int = 255036"},{name:"alignment_intermediate_size",val:": int = 36864"},{name:"tie_word_embeddings",val:": bool = True"}],parametersDescription:[{anchor:"transformers.Cohere2VisionConfig.vision_config",description:`<strong>vision_config</strong> (<code>Union[dict, ~configuration_utils.PreTrainedConfig]</code>, <em>optional</em>) &#x2014;
The config object or dictionary of the vision backbone.`,name:"vision_config"},{anchor:"transformers.Cohere2VisionConfig.text_config",description:`<strong>text_config</strong> (<code>Union[dict, ~configuration_utils.PreTrainedConfig]</code>, <em>optional</em>) &#x2014;
The config object or dictionary of the text backbone.`,name:"text_config"},{anchor:"transformers.Cohere2VisionConfig.downsample_factor",description:`<strong>downsample_factor</strong> (<code>int</code>, <em>optional</em>, defaults to 2) &#x2014;
The factor by which to downsample the input image.`,name:"downsample_factor"},{anchor:"transformers.Cohere2VisionConfig.image_token_id",description:`<strong>image_token_id</strong> (<code>int</code>, <em>optional</em>, defaults to <code>255036</code>) &#x2014;
The image token index used as a placeholder for input images.`,name:"image_token_id"},{anchor:"transformers.Cohere2VisionConfig.alignment_intermediate_size",description:`<strong>alignment_intermediate_size</strong> (<code>int</code>, <em>optional</em>, defaults to 36864) &#x2014;
The size of the intermediate layer for alignment.`,name:"alignment_intermediate_size"},{anchor:"transformers.Cohere2VisionConfig.tie_word_embeddings",description:`<strong>tie_word_embeddings</strong> (<code>bool</code>, <em>optional</em>, defaults to <code>True</code>) &#x2014;
Whether to tie weight embeddings according to model&#x2019;s <code>tied_weights_keys</code> mapping.`,name:"tie_word_embeddings"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/configuration_cohere2_vision.py#L25"}}),de=new K({props:{title:"Cohere2VisionForConditionalGeneration",local:"transformers.Cohere2VisionForConditionalGeneration",headingTag:"h2"}}),pe=new A({props:{name:"class transformers.Cohere2VisionForConditionalGeneration",anchor:"transformers.Cohere2VisionForConditionalGeneration",parameters:[{name:"config",val:": Cohere2VisionConfig"}],parametersDescription:[{anchor:"transformers.Cohere2VisionForConditionalGeneration.config",description:`<strong>config</strong> (<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionConfig">Cohere2VisionConfig</a>) &#x2014;
Model configuration class with all the parameters of the model. Initializing with a config file does not
load the weights associated with the model, only the configuration. Check out the
<a href="/docs/transformers/pr_43265/en/main_classes/model#transformers.PreTrainedModel.from_pretrained">from_pretrained()</a> method to load the model weights.`,name:"config"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/modeling_cohere2_vision.py#L242"}}),me=new A({props:{name:"forward",anchor:"transformers.Cohere2VisionForConditionalGeneration.forward",parameters:[{name:"input_ids",val:": torch.LongTensor | None = None"},{name:"pixel_values",val:": torch.FloatTensor | None = None"},{name:"attention_mask",val:": torch.Tensor | None = None"},{name:"position_ids",val:": torch.LongTensor | None = None"},{name:"past_key_values",val:": transformers.cache_utils.Cache | None = None"},{name:"inputs_embeds",val:": torch.FloatTensor | None = None"},{name:"labels",val:": torch.LongTensor | None = None"},{name:"use_cache",val:": bool | None = None"},{name:"logits_to_keep",val:": int | torch.Tensor = 0"},{name:"image_sizes",val:": torch.Tensor | None = None"},{name:"**kwargs",val:": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}],parametersDescription:[{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.input_ids",description:`<strong>input_ids</strong> (<code>torch.LongTensor</code> of shape <code>(batch_size, sequence_length)</code>, <em>optional</em>) &#x2014;
Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.</p>
<p>Indices can be obtained using <a href="/docs/transformers/pr_43265/en/model_doc/auto#transformers.AutoTokenizer">AutoTokenizer</a>. See <a href="/docs/transformers/pr_43265/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode">PreTrainedTokenizer.encode()</a> and
<a href="/docs/transformers/pr_43265/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__">PreTrainedTokenizer.<strong>call</strong>()</a> for details.</p>
<p><a href="../glossary#input-ids">What are input IDs?</a>`,name:"input_ids"},{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.pixel_values",description:`<strong>pixel_values</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, num_channels, image_size, image_size)</code>, <em>optional</em>) &#x2014;
The tensors corresponding to the input images. Pixel values can be obtained using
<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">Cohere2VisionImageProcessor</a>. See <code>Cohere2VisionImageProcessor.__call__()</code> for details (<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionProcessor">Cohere2VisionProcessor</a> uses
<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">Cohere2VisionImageProcessor</a> for processing images).`,name:"pixel_values"},{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.attention_mask",description:`<strong>attention_mask</strong> (<code>torch.Tensor</code> of shape <code>(batch_size, sequence_length)</code>, <em>optional</em>) &#x2014;
Mask to avoid performing attention on padding token indices. Mask values selected in <code>[0, 1]</code>:</p>
<ul>
<li>1 for tokens that are <strong>not masked</strong>,</li>
<li>0 for tokens that are <strong>masked</strong>.</li>
</ul>
<p><a href="../glossary#attention-mask">What are attention masks?</a>`,name:"attention_mask"},{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.position_ids",description:`<strong>position_ids</strong> (<code>torch.LongTensor</code> of shape <code>(batch_size, sequence_length)</code>, <em>optional</em>) &#x2014;
Indices of positions of each input sequence tokens in the position embeddings. Selected in the range <code>[0, config.n_positions - 1]</code>.</p>
<p><a href="../glossary#position-ids">What are position IDs?</a>`,name:"position_ids"},{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.past_key_values",description:`<strong>past_key_values</strong> (<code>~cache_utils.Cache</code>, <em>optional</em>) &#x2014;
Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
blocks) that can be used to speed up sequential decoding. This typically consists in the <code>past_key_values</code>
returned by the model at a previous stage of decoding, when <code>use_cache=True</code> or <code>config.use_cache=True</code>.</p>
<p>Only <a href="/docs/transformers/pr_43265/en/internal/generation_utils#transformers.Cache">Cache</a> instance is allowed as input, see our <a href="https://huggingface.co/docs/transformers/en/kv_cache" rel="nofollow">kv cache guide</a>.
If no <code>past_key_values</code> are passed, <a href="/docs/transformers/pr_43265/en/internal/generation_utils#transformers.DynamicCache">DynamicCache</a> will be initialized by default.</p>
<p>The model will output the same cache format that is fed as input.</p>
<p>If <code>past_key_values</code> are used, the user is expected to input only unprocessed <code>input_ids</code> (those that don&#x2019;t
have their past key value states given to this model) of shape <code>(batch_size, unprocessed_length)</code> instead of all <code>input_ids</code>
of shape <code>(batch_size, sequence_length)</code>.`,name:"past_key_values"},{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.inputs_embeds",description:`<strong>inputs_embeds</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, sequence_length, hidden_size)</code>, <em>optional</em>) &#x2014;
Optionally, instead of passing <code>input_ids</code> you can choose to directly pass an embedded representation. This
is useful if you want more control over how to convert <code>input_ids</code> indices into associated vectors than the
model&#x2019;s internal embedding lookup matrix.`,name:"inputs_embeds"},{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.labels",description:`<strong>labels</strong> (<code>torch.LongTensor</code> of shape <code>(batch_size, sequence_length)</code>, <em>optional</em>) &#x2014;
Labels for computing the masked language modeling loss. Indices should either be in <code>[0, ..., config.vocab_size]</code> or -100 (see <code>input_ids</code> docstring). Tokens with indices set to <code>-100</code> are ignored
(masked), the loss is only computed for the tokens with labels in <code>[0, ..., config.vocab_size]</code>.`,name:"labels"},{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.use_cache",description:`<strong>use_cache</strong> (<code>bool</code>, <em>optional</em>) &#x2014;
If set to <code>True</code>, <code>past_key_values</code> key value states are returned and can be used to speed up decoding (see
<code>past_key_values</code>).`,name:"use_cache"},{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.logits_to_keep",description:`<strong>logits_to_keep</strong> (<code>Union[int, torch.Tensor]</code>, <em>optional</em>, defaults to <code>0</code>) &#x2014;
If an <code>int</code>, compute logits for the last <code>logits_to_keep</code> tokens. If <code>0</code>, calculate logits for all
<code>input_ids</code> (special case). Only last token logits are needed for generation, and calculating them only for that
token can save memory, which becomes pretty significant for long sequences or large vocabulary size.
If a <code>torch.Tensor</code>, must be 1D corresponding to the indices to keep in the sequence length dimension.
This is useful when using packed tensor format (single dimension for batch and sequence length).`,name:"logits_to_keep"},{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.image_sizes",description:`<strong>image_sizes</strong> (<code>torch.Tensor</code> of shape <code>(batch_size, 2)</code>, <em>optional</em>) &#x2014;
The sizes of the images in the batch, being (height, width) for each image.`,name:"image_sizes"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/modeling_cohere2_vision.py#L260",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p>A <code>Cohere2VisionCausalLMOutputWithPast</code> or a tuple of
<code>torch.FloatTensor</code> (if <code>return_dict=False</code> is passed or when <code>config.return_dict=False</code>) comprising various
elements depending on the configuration (<a
href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionConfig"
>Cohere2VisionConfig</a>) and inputs.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><code>Cohere2VisionCausalLMOutputWithPast</code> or <code>tuple(torch.FloatTensor)</code></p>
`}}),L=new Mt({props:{$$slots:{default:[xt]},$$scope:{ctx:U}}}),H=new _o({props:{anchor:"transformers.Cohere2VisionForConditionalGeneration.forward.example",$$slots:{default:[zt]},$$scope:{ctx:U}}}),he=new A({props:{name:"get_image_features",anchor:"transformers.Cohere2VisionForConditionalGeneration.get_image_features",parameters:[{name:"pixel_values",val:": FloatTensor"},{name:"**kwargs",val:": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}],parametersDescription:[{anchor:"transformers.Cohere2VisionForConditionalGeneration.get_image_features.pixel_values",description:`<strong>pixel_values</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, num_channels, image_size, image_size)</code>) &#x2014;
The tensors corresponding to the input images. Pixel values can be obtained using
<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">Cohere2VisionImageProcessor</a>. See <code>Cohere2VisionImageProcessor.__call__()</code> for details (<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionProcessor">Cohere2VisionProcessor</a> uses
<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">Cohere2VisionImageProcessor</a> for processing images).`,name:"pixel_values"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/modeling_cohere2_vision.py#L254",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p>A <a
href="/docs/transformers/pr_43265/en/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling"
>BaseModelOutputWithPooling</a> or a tuple of
<code>torch.FloatTensor</code> (if <code>return_dict=False</code> is passed or when <code>config.return_dict=False</code>) comprising various
elements depending on the configuration (<a
href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionConfig"
>Cohere2VisionConfig</a>) and inputs.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><a
href="/docs/transformers/pr_43265/en/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling"
>BaseModelOutputWithPooling</a> or <code>tuple(torch.FloatTensor)</code></p>
`}}),X=new _o({props:{anchor:"transformers.Cohere2VisionForConditionalGeneration.get_image_features.example",$$slots:{default:[Ft]},$$scope:{ctx:U}}}),ue=new K({props:{title:"Cohere2VisionModel",local:"transformers.Cohere2VisionModel",headingTag:"h2"}}),ge=new A({props:{name:"class transformers.Cohere2VisionModel",anchor:"transformers.Cohere2VisionModel",parameters:[{name:"config",val:": Cohere2VisionConfig"}],parametersDescription:[{anchor:"transformers.Cohere2VisionModel.config",description:`<strong>config</strong> (<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionConfig">Cohere2VisionConfig</a>) &#x2014;
Model configuration class with all the parameters of the model. Initializing with a config file does not
load the weights associated with the model, only the configuration. Check out the
<a href="/docs/transformers/pr_43265/en/main_classes/model#transformers.PreTrainedModel.from_pretrained">from_pretrained()</a> method to load the model weights.`,name:"config"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/modeling_cohere2_vision.py#L146"}}),fe=new A({props:{name:"forward",anchor:"transformers.Cohere2VisionModel.forward",parameters:[{name:"input_ids",val:": torch.LongTensor | None = None"},{name:"pixel_values",val:": torch.FloatTensor | None = None"},{name:"attention_mask",val:": torch.Tensor | None = None"},{name:"position_ids",val:": torch.LongTensor | None = None"},{name:"past_key_values",val:": transformers.cache_utils.Cache | None = None"},{name:"inputs_embeds",val:": torch.FloatTensor | None = None"},{name:"use_cache",val:": bool | None = None"},{name:"**kwargs",val:": typing_extensions.Unpack[transformers.modeling_flash_attention_utils.FlashAttentionKwargs]"}],parametersDescription:[{anchor:"transformers.Cohere2VisionModel.forward.input_ids",description:`<strong>input_ids</strong> (<code>torch.LongTensor</code> of shape <code>(batch_size, sequence_length)</code>, <em>optional</em>) &#x2014;
Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.</p>
<p>Indices can be obtained using <a href="/docs/transformers/pr_43265/en/model_doc/auto#transformers.AutoTokenizer">AutoTokenizer</a>. See <a href="/docs/transformers/pr_43265/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode">PreTrainedTokenizer.encode()</a> and
<a href="/docs/transformers/pr_43265/en/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__">PreTrainedTokenizer.<strong>call</strong>()</a> for details.</p>
<p><a href="../glossary#input-ids">What are input IDs?</a>`,name:"input_ids"},{anchor:"transformers.Cohere2VisionModel.forward.pixel_values",description:`<strong>pixel_values</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, num_channels, image_size, image_size)</code>, <em>optional</em>) &#x2014;
The tensors corresponding to the input images. Pixel values can be obtained using
<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">Cohere2VisionImageProcessor</a>. See <code>Cohere2VisionImageProcessor.__call__()</code> for details (<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionProcessor">Cohere2VisionProcessor</a> uses
<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">Cohere2VisionImageProcessor</a> for processing images).`,name:"pixel_values"},{anchor:"transformers.Cohere2VisionModel.forward.attention_mask",description:`<strong>attention_mask</strong> (<code>torch.Tensor</code> of shape <code>(batch_size, sequence_length)</code>, <em>optional</em>) &#x2014;
Mask to avoid performing attention on padding token indices. Mask values selected in <code>[0, 1]</code>:</p>
<ul>
<li>1 for tokens that are <strong>not masked</strong>,</li>
<li>0 for tokens that are <strong>masked</strong>.</li>
</ul>
<p><a href="../glossary#attention-mask">What are attention masks?</a>`,name:"attention_mask"},{anchor:"transformers.Cohere2VisionModel.forward.position_ids",description:`<strong>position_ids</strong> (<code>torch.LongTensor</code> of shape <code>(batch_size, sequence_length)</code>, <em>optional</em>) &#x2014;
Indices of positions of each input sequence tokens in the position embeddings. Selected in the range <code>[0, config.n_positions - 1]</code>.</p>
<p><a href="../glossary#position-ids">What are position IDs?</a>`,name:"position_ids"},{anchor:"transformers.Cohere2VisionModel.forward.past_key_values",description:`<strong>past_key_values</strong> (<code>~cache_utils.Cache</code>, <em>optional</em>) &#x2014;
Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
blocks) that can be used to speed up sequential decoding. This typically consists in the <code>past_key_values</code>
returned by the model at a previous stage of decoding, when <code>use_cache=True</code> or <code>config.use_cache=True</code>.</p>
<p>Only <a href="/docs/transformers/pr_43265/en/internal/generation_utils#transformers.Cache">Cache</a> instance is allowed as input, see our <a href="https://huggingface.co/docs/transformers/en/kv_cache" rel="nofollow">kv cache guide</a>.
If no <code>past_key_values</code> are passed, <a href="/docs/transformers/pr_43265/en/internal/generation_utils#transformers.DynamicCache">DynamicCache</a> will be initialized by default.</p>
<p>The model will output the same cache format that is fed as input.</p>
<p>If <code>past_key_values</code> are used, the user is expected to input only unprocessed <code>input_ids</code> (those that don&#x2019;t
have their past key value states given to this model) of shape <code>(batch_size, unprocessed_length)</code> instead of all <code>input_ids</code>
of shape <code>(batch_size, sequence_length)</code>.`,name:"past_key_values"},{anchor:"transformers.Cohere2VisionModel.forward.inputs_embeds",description:`<strong>inputs_embeds</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, sequence_length, hidden_size)</code>, <em>optional</em>) &#x2014;
Optionally, instead of passing <code>input_ids</code> you can choose to directly pass an embedded representation. This
is useful if you want more control over how to convert <code>input_ids</code> indices into associated vectors than the
model&#x2019;s internal embedding lookup matrix.`,name:"inputs_embeds"},{anchor:"transformers.Cohere2VisionModel.forward.use_cache",description:`<strong>use_cache</strong> (<code>bool</code>, <em>optional</em>) &#x2014;
If set to <code>True</code>, <code>past_key_values</code> key value states are returned and can be used to speed up decoding (see
<code>past_key_values</code>).`,name:"use_cache"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/modeling_cohere2_vision.py#L192",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p>A <code>Cohere2VisionModelOutputWithPast</code> or a tuple of
<code>torch.FloatTensor</code> (if <code>return_dict=False</code> is passed or when <code>config.return_dict=False</code>) comprising various
elements depending on the configuration (<a
href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionConfig"
>Cohere2VisionConfig</a>) and inputs.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><code>Cohere2VisionModelOutputWithPast</code> or <code>tuple(torch.FloatTensor)</code></p>
`}}),S=new Mt({props:{$$slots:{default:[Bt]},$$scope:{ctx:U}}}),D=new _o({props:{anchor:"transformers.Cohere2VisionModel.forward.example",$$slots:{default:[qt]},$$scope:{ctx:U}}}),_e=new A({props:{name:"get_image_features",anchor:"transformers.Cohere2VisionModel.get_image_features",parameters:[{name:"pixel_values",val:": FloatTensor"},{name:"**kwargs",val:": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}],parametersDescription:[{anchor:"transformers.Cohere2VisionModel.get_image_features.pixel_values",description:`<strong>pixel_values</strong> (<code>torch.FloatTensor</code> of shape <code>(batch_size, num_channels, image_size, image_size)</code>) &#x2014;
The tensors corresponding to the input images. Pixel values can be obtained using
<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">Cohere2VisionImageProcessor</a>. See <code>Cohere2VisionImageProcessor.__call__()</code> for details (<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionProcessor">Cohere2VisionProcessor</a> uses
<a href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionImageProcessor">Cohere2VisionImageProcessor</a> for processing images).`,name:"pixel_values"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/modeling_cohere2_vision.py#L155",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p>A <a
href="/docs/transformers/pr_43265/en/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling"
>BaseModelOutputWithPooling</a> or a tuple of
<code>torch.FloatTensor</code> (if <code>return_dict=False</code> is passed or when <code>config.return_dict=False</code>) comprising various
elements depending on the configuration (<a
href="/docs/transformers/pr_43265/en/model_doc/cohere2_vision#transformers.Cohere2VisionConfig"
>Cohere2VisionConfig</a>) and inputs.</p>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><a
href="/docs/transformers/pr_43265/en/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling"
>BaseModelOutputWithPooling</a> or <code>tuple(torch.FloatTensor)</code></p>
`}}),Y=new _o({props:{anchor:"transformers.Cohere2VisionModel.get_image_features.example",$$slots:{default:[Pt]},$$scope:{ctx:U}}}),ye=new K({props:{title:"Cohere2VisionImageProcessor",local:"transformers.Cohere2VisionImageProcessor",headingTag:"h2"}}),Me=new A({props:{name:"class transformers.Cohere2VisionImageProcessor",anchor:"transformers.Cohere2VisionImageProcessor",parameters:[{name:"**kwargs",val:": typing_extensions.Unpack[transformers.models.cohere2_vision.image_processing_cohere2_vision.Cohere2VisionImageProcessorKwargs]"}],parametersDescription:[{anchor:"transformers.Cohere2VisionImageProcessor.crop_to_patches",description:`<strong>crop_to_patches</strong> (<code>bool</code>, <em>kwargs</em>, <em>optional</em>, defaults to <code>False</code>) &#x2014;
Whether to crop the image to patches. Can be overridden by the <code>crop_to_patches</code> parameter in the
<code>preprocess</code> method.`,name:"crop_to_patches"},{anchor:"transformers.Cohere2VisionImageProcessor.min_patches",description:`<strong>min_patches</strong> (<code>int</code>, <em>kwargs</em>, <em>optional</em>, defaults to 1) &#x2014;
The minimum number of patches to be extracted from the image. Only has an effect if <code>crop_to_patches</code> is
set to <code>True</code>. Can be overridden by the <code>min_patches</code> parameter in the <code>preprocess</code> method.`,name:"min_patches"},{anchor:"transformers.Cohere2VisionImageProcessor.max_patches",description:`<strong>max_patches</strong> (<code>int</code>, <em>kwargs</em>, <em>optional</em>, defaults to 12) &#x2014;
The maximum number of patches to be extracted from the image. Only has an effect if <code>crop_to_patches</code> is
set to <code>True</code>. Can be overridden by the <code>max_patches</code> parameter in the <code>preprocess</code> method.`,name:"max_patches"},{anchor:"transformers.Cohere2VisionImageProcessor.*kwargs",description:`*<strong>*kwargs</strong> (<a href="/docs/transformers/pr_43265/en/main_classes/processors#transformers.ImagesKwargs">ImagesKwargs</a>, <em>optional</em>) &#x2014;
Additional image preprocessing options. Model-specific kwargs are listed above; see the TypedDict class
for the complete list of supported arguments.`,name:"*kwargs"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/image_processing_cohere2_vision.py#L110"}}),Te=new A({props:{name:"preprocess",anchor:"transformers.Cohere2VisionImageProcessor.preprocess",parameters:[{name:"images",val:": typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), list['PIL.Image.Image'], list[numpy.ndarray], list['torch.Tensor']]"},{name:"*args",val:""},{name:"**kwargs",val:": typing_extensions.Unpack[transformers.processing_utils.ImagesKwargs]"}],parametersDescription:[{anchor:"transformers.Cohere2VisionImageProcessor.preprocess.images",description:`<strong>images</strong> (<code>Union[PIL.Image.Image, numpy.ndarray, torch.Tensor, list[PIL.Image.Image], list[numpy.ndarray], list[torch.Tensor]]</code>) &#x2014;
Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If
passing in images with pixel values between 0 and 1, set <code>do_rescale=False</code>.`,name:"images"},{anchor:"transformers.Cohere2VisionImageProcessor.preprocess.return_tensors",description:`<strong>return_tensors</strong> (<code>str</code> or <a href="/docs/transformers/pr_43265/en/internal/file_utils#transformers.TensorType">TensorType</a>, <em>optional</em>) &#x2014;
Returns stacked tensors if set to <code>&apos;pt&apos;</code>, otherwise returns a list of tensors.`,name:"return_tensors"},{anchor:"transformers.Cohere2VisionImageProcessor.preprocess.*kwargs",description:`*<strong>*kwargs</strong> (<a href="/docs/transformers/pr_43265/en/main_classes/processors#transformers.ImagesKwargs">ImagesKwargs</a>, <em>optional</em>) &#x2014;
Additional image preprocessing options. Model-specific kwargs are listed above; see the TypedDict class
for the complete list of supported arguments.`,name:"*kwargs"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/image_processing_utils.py#L382",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<ul>
<li><strong>data</strong> (<code>dict</code>) — Dictionary of lists/arrays/tensors returned by the <strong>call</strong> method (‘pixel_values’, etc.).</li>
<li><strong>tensor_type</strong> (<code>Union[None, str, TensorType]</code>, <em>optional</em>) — You can give a tensor_type here to convert the lists of integers in PyTorch/Numpy Tensors at
initialization.</li>
</ul>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><code>~image_processing_base.BatchFeature</code></p>
`}}),Ce=new K({props:{title:"Cohere2VisionProcessor",local:"transformers.Cohere2VisionProcessor",headingTag:"h2"}}),we=new A({props:{name:"class transformers.Cohere2VisionProcessor",anchor:"transformers.Cohere2VisionProcessor",parameters:[{name:"image_processor",val:" = None"},{name:"tokenizer",val:" = None"},{name:"chat_template",val:" = None"},{name:"**kwargs",val:""}],parametersDescription:[{anchor:"transformers.Cohere2VisionProcessor.image_processor",description:`<strong>image_processor</strong> (<code>Cohere2VisionImageProcessor</code>) &#x2014;
The image processor is a required input.`,name:"image_processor"},{anchor:"transformers.Cohere2VisionProcessor.tokenizer",description:`<strong>tokenizer</strong> (<code>tokenizer_class</code>) &#x2014;
The tokenizer is a required input.`,name:"tokenizer"},{anchor:"transformers.Cohere2VisionProcessor.chat_template",description:`<strong>chat_template</strong> (<code>str</code>) &#x2014;
A Jinja template to convert lists of messages in a chat into a tokenizable string.`,name:"chat_template"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/processing_cohere2_vision.py#L34"}}),Ue=new A({props:{name:"__call__",anchor:"transformers.Cohere2VisionProcessor.__call__",parameters:[{name:"images",val:": typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), list['PIL.Image.Image'], list[numpy.ndarray], list['torch.Tensor'], NoneType] = None"},{name:"text",val:": str | list[str] | list[list[str]] | None = None"},{name:"**kwargs",val:": typing_extensions.Unpack[transformers.models.cohere2_vision.processing_cohere2_vision.Cohere2VisionProcessorKwargs]"}],parametersDescription:[{anchor:"transformers.Cohere2VisionProcessor.__call__.images",description:`<strong>images</strong> (<code>Union[PIL.Image.Image, numpy.ndarray, torch.Tensor, list[PIL.Image.Image], list[numpy.ndarray], list[torch.Tensor]]</code>, <em>optional</em>) &#x2014;
Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If
passing in images with pixel values between 0 and 1, set <code>do_rescale=False</code>.`,name:"images"},{anchor:"transformers.Cohere2VisionProcessor.__call__.text",description:`<strong>text</strong> (<code>Union[str, list[str], list[list[str]]]</code>, <em>optional</em>) &#x2014;
The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings
(pretokenized string). If you pass a pretokenized input, set <code>is_split_into_words=True</code> to avoid ambiguity with batched inputs.`,name:"text"},{anchor:"transformers.Cohere2VisionProcessor.__call__.return_tensors",description:`<strong>return_tensors</strong> (<code>str</code> or <a href="/docs/transformers/pr_43265/en/internal/file_utils#transformers.TensorType">TensorType</a>, <em>optional</em>) &#x2014;
If set, will return tensors of a particular framework. Acceptable values are:</p>
<ul>
<li><code>&apos;pt&apos;</code>: Return PyTorch <code>torch.Tensor</code> objects.</li>
<li><code>&apos;np&apos;</code>: Return NumPy <code>np.ndarray</code> objects.</li>
</ul>`,name:"return_tensors"},{anchor:"transformers.Cohere2VisionProcessor.__call__.*kwargs",description:`*<strong>*kwargs</strong> (<a href="/docs/transformers/pr_43265/en/main_classes/processors#transformers.ProcessingKwargs">ProcessingKwargs</a>, <em>optional</em>) &#x2014;
Additional processing options for each modality (text, images, videos, audio). Model-specific parameters
are listed above; see the TypedDict class for the complete list of supported arguments.`,name:"*kwargs"}],source:"https://github.com/huggingface/transformers/blob/vr_43265/src/transformers/models/cohere2_vision/processing_cohere2_vision.py#L62",returnDescription:`<script context="module">export const metadata = 'undefined';<\/script>
<p>A <a
href="/docs/transformers/pr_43265/en/main_classes/feature_extractor#transformers.BatchFeature"
>BatchFeature</a> with the following fields:</p>
<ul>
<li><strong>input_ids</strong> — List of token ids to be fed to a model. Returned when <code>text</code> is not <code>None</code>.</li>
<li><strong>attention_mask</strong> — List of indices specifying which tokens should be attended to by the model (when
<code>return_attention_mask=True</code> or if <em>“attention_mask”</em> is in <code>self.model_input_names</code> and if <code>text</code> is not
<code>None</code>).</li>
<li><strong>pixel_values</strong> — Pixel values to be fed to a model. Returned when <code>images</code> is not <code>None</code>.</li>
</ul>
`,returnType:`<script context="module">export const metadata = 'undefined';<\/script>
<p><a
href="/docs/transformers/pr_43265/en/main_classes/feature_extractor#transformers.BatchFeature"
>BatchFeature</a></p>
`}}),be=new It({props:{source:"https://github.com/huggingface/transformers/blob/main/docs/source/en/model_doc/cohere2_vision.md"}}),{c(){o=m("meta"),u=a(),s=m("p"),c=a(),d=m("p"),d.innerHTML=t,w=a(),g(N.$$.fragment),Xe=a(),g(oe.$$.fragment),Se=a(),R=m("div"),R.innerHTML=Xo,De=a(),g(te.$$.fragment),Ye=a(),se=m("p"),se.innerHTML=So,Oe=a(),ne=m("p"),ne.textContent=Do,Ke=a(),re=m("p"),re.textContent=Yo,eo=a(),g(ae.$$.fragment),oo=a(),ie=m("p"),ie.textContent=Oo,to=a(),g(Q.$$.fragment),so=a(),g(le.$$.fragment),no=a(),q=m("div"),g(ce.$$.fragment),yo=a(),ve=m("p"),ve.innerHTML=Ko,Mo=a(),Je=m("p"),Je.innerHTML=et,ro=a(),g(de.$$.fragment),ao=a(),b=m("div"),g(pe.$$.fragment),To=a(),Ie=m("p"),Ie.textContent=ot,Co=a(),je=m("p"),je.innerHTML=tt,wo=a(),$e=m("p"),$e.innerHTML=st,Uo=a(),j=m("div"),g(me.$$.fragment),bo=a(),ke=m("p"),ke.innerHTML=nt,vo=a(),g(L.$$.fragment),Jo=a(),Ve=m("ul"),Ve.innerHTML=rt,Io=a(),g(H.$$.fragment),jo=a(),G=m("div"),g(he.$$.fragment),$o=a(),xe=m("ul"),xe.innerHTML=at,ko=a(),g(X.$$.fragment),io=a(),g(ue.$$.fragment),lo=a(),v=m("div"),g(ge.$$.fragment),Vo=a(),ze=m("p"),ze.textContent=it,xo=a(),Fe=m("p"),Fe.innerHTML=lt,zo=a(),Be=m("p"),Be.innerHTML=ct,Fo=a(),$=m("div"),g(fe.$$.fragment),Bo=a(),qe=m("p"),qe.innerHTML=dt,qo=a(),g(S.$$.fragment),Po=a(),Pe=m("ul"),Pe.innerHTML=pt,Zo=a(),g(D.$$.fragment),Ao=a(),B=m("div"),g(_e.$$.fragment),No=a(),Ze=m("p"),Ze.textContent=mt,Go=a(),Ae=m("ul"),Ae.innerHTML=ht,Eo=a(),g(Y.$$.fragment),co=a(),g(ye.$$.fragment),po=a(),P=m("div"),g(Me.$$.fragment),Wo=a(),Ne=m("p"),Ne.textContent=ut,Ro=a(),Ge=m("div"),g(Te.$$.fragment),mo=a(),g(Ce.$$.fragment),ho=a(),V=m("div"),g(we.$$.fragment),Qo=a(),Ee=m("p"),Ee.textContent=gt,Lo=a(),We=m("p"),We.innerHTML=ft,Ho=a(),Re=m("div"),g(Ue.$$.fragment),uo=a(),g(be.$$.fragment),go=a(),He=m("p"),this.h()},l(e){const n=vt("svelte-u9bgzb",document.head);o=h(n,"META",{name:!0,content:!0}),n.forEach(r),u=i(e),s=h(e,"P",{}),F(s).forEach(r),c=i(e),d=h(e,"P",{"data-svelte-h":!0}),C(d)!=="svelte-dcyqwm"&&(d.innerHTML=t),w=i(e),f(N.$$.fragment,e),Xe=i(e),f(oe.$$.fragment,e),Se=i(e),R=h(e,"DIV",{class:!0,"data-svelte-h":!0}),C(R)!=="svelte-1iyo9lh"&&(R.innerHTML=Xo),De=i(e),f(te.$$.fragment,e),Ye=i(e),se=h(e,"P",{"data-svelte-h":!0}),C(se)!=="svelte-mp5n58"&&(se.innerHTML=So),Oe=i(e),ne=h(e,"P",{"data-svelte-h":!0}),C(ne)!=="svelte-18ri1fu"&&(ne.textContent=Do),Ke=i(e),re=h(e,"P",{"data-svelte-h":!0}),C(re)!=="svelte-2q66tf"&&(re.textContent=Yo),eo=i(e),f(ae.$$.fragment,e),oo=i(e),ie=h(e,"P",{"data-svelte-h":!0}),C(ie)!=="svelte-1a3wb26"&&(ie.textContent=Oo),to=i(e),f(Q.$$.fragment,e),so=i(e),f(le.$$.fragment,e),no=i(e),q=h(e,"DIV",{class:!0});var E=F(q);f(ce.$$.fragment,E),yo=i(E),ve=h(E,"P",{"data-svelte-h":!0}),C(ve)!=="svelte-dk9sv8"&&(ve.innerHTML=Ko),Mo=i(E),Je=h(E,"P",{"data-svelte-h":!0}),C(Je)!=="svelte-1e8815j"&&(Je.innerHTML=et),E.forEach(r),ro=i(e),f(de.$$.fragment,e),ao=i(e),b=h(e,"DIV",{class:!0});var J=F(b);f(pe.$$.fragment,J),To=i(J),Ie=h(J,"P",{"data-svelte-h":!0}),C(Ie)!=="svelte-hiv3do"&&(Ie.textContent=ot),Co=i(J),je=h(J,"P",{"data-svelte-h":!0}),C(je)!=="svelte-1gb3c10"&&(je.innerHTML=tt),wo=i(J),$e=h(J,"P",{"data-svelte-h":!0}),C($e)!=="svelte-hswkmf"&&($e.innerHTML=st),Uo=i(J),j=h(J,"DIV",{class:!0});var x=F(j);f(me.$$.fragment,x),bo=i(x),ke=h(x,"P",{"data-svelte-h":!0}),C(ke)!=="svelte-8p4026"&&(ke.innerHTML=nt),vo=i(x),f(L.$$.fragment,x),Jo=i(x),Ve=h(x,"UL",{"data-svelte-h":!0}),C(Ve)!=="svelte-1iewcof"&&(Ve.innerHTML=rt),Io=i(x),f(H.$$.fragment,x),x.forEach(r),jo=i(J),G=h(J,"DIV",{class:!0});var W=F(G);f(he.$$.fragment,W),$o=i(W),xe=h(W,"UL",{"data-svelte-h":!0}),C(xe)!=="svelte-1vt4ztn"&&(xe.innerHTML=at),ko=i(W),f(X.$$.fragment,W),W.forEach(r),J.forEach(r),io=i(e),f(ue.$$.fragment,e),lo=i(e),v=h(e,"DIV",{class:!0});var I=F(v);f(ge.$$.fragment,I),Vo=i(I),ze=h(I,"P",{"data-svelte-h":!0}),C(ze)!=="svelte-1kgnlur"&&(ze.textContent=it),xo=i(I),Fe=h(I,"P",{"data-svelte-h":!0}),C(Fe)!=="svelte-1gb3c10"&&(Fe.innerHTML=lt),zo=i(I),Be=h(I,"P",{"data-svelte-h":!0}),C(Be)!=="svelte-hswkmf"&&(Be.innerHTML=ct),Fo=i(I),$=h(I,"DIV",{class:!0});var z=F($);f(fe.$$.fragment,z),Bo=i(z),qe=h(z,"P",{"data-svelte-h":!0}),C(qe)!=="svelte-k6k5h8"&&(qe.innerHTML=dt),qo=i(z),f(S.$$.fragment,z),Po=i(z),Pe=h(z,"UL",{"data-svelte-h":!0}),C(Pe)!=="svelte-1boxntb"&&(Pe.innerHTML=pt),Zo=i(z),f(D.$$.fragment,z),z.forEach(r),Ao=i(I),B=h(I,"DIV",{class:!0});var Z=F(B);f(_e.$$.fragment,Z),No=i(Z),Ze=h(Z,"P",{"data-svelte-h":!0}),C(Ze)!=="svelte-1vzo9k5"&&(Ze.textContent=mt),Go=i(Z),Ae=h(Z,"UL",{"data-svelte-h":!0}),C(Ae)!=="svelte-1vt4ztn"&&(Ae.innerHTML=ht),Eo=i(Z),f(Y.$$.fragment,Z),Z.forEach(r),I.forEach(r),co=i(e),f(ye.$$.fragment,e),po=i(e),P=h(e,"DIV",{class:!0});var Qe=F(P);f(Me.$$.fragment,Qe),Wo=i(Qe),Ne=h(Qe,"P",{"data-svelte-h":!0}),C(Ne)!=="svelte-13trhbl"&&(Ne.textContent=ut),Ro=i(Qe),Ge=h(Qe,"DIV",{class:!0});var _t=F(Ge);f(Te.$$.fragment,_t),_t.forEach(r),Qe.forEach(r),mo=i(e),f(Ce.$$.fragment,e),ho=i(e),V=h(e,"DIV",{class:!0});var O=F(V);f(we.$$.fragment,O),Qo=i(O),Ee=h(O,"P",{"data-svelte-h":!0}),C(Ee)!=="svelte-p24co5"&&(Ee.textContent=gt),Lo=i(O),We=h(O,"P",{"data-svelte-h":!0}),C(We)!=="svelte-127hg95"&&(We.innerHTML=ft),Ho=i(O),Re=h(O,"DIV",{class:!0});var yt=F(Re);f(Ue.$$.fragment,yt),yt.forEach(r),O.forEach(r),uo=i(e),f(be.$$.fragment,e),go=i(e),He=h(e,"P",{}),F(He).forEach(r),this.h()},h(){k(o,"name","hf:doc:metadata"),k(o,"content",At),k(R,"class","flex flex-wrap space-x-1"),k(q,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k(j,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k(G,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k(b,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k($,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k(B,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k(v,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k(Ge,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k(P,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k(Re,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8"),k(V,"class","docstring border-l-2 border-t-2 pl-4 pt-3.5 border-gray-100 rounded-tl-xl mb-6 mt-8")},m(e,n){l(document.head,o),p(e,u,n),p(e,s,n),p(e,c,n),p(e,d,n),p(e,w,n),_(N,e,n),p(e,Xe,n),_(oe,e,n),p(e,Se,n),p(e,R,n),p(e,De,n),_(te,e,n),p(e,Ye,n),p(e,se,n),p(e,Oe,n),p(e,ne,n),p(e,Ke,n),p(e,re,n),p(e,eo,n),_(ae,e,n),p(e,oo,n),p(e,ie,n),p(e,to,n),_(Q,e,n),p(e,so,n),_(le,e,n),p(e,no,n),p(e,q,n),_(ce,q,null),l(q,yo),l(q,ve),l(q,Mo),l(q,Je),p(e,ro,n),_(de,e,n),p(e,ao,n),p(e,b,n),_(pe,b,null),l(b,To),l(b,Ie),l(b,Co),l(b,je),l(b,wo),l(b,$e),l(b,Uo),l(b,j),_(me,j,null),l(j,bo),l(j,ke),l(j,vo),_(L,j,null),l(j,Jo),l(j,Ve),l(j,Io),_(H,j,null),l(b,jo),l(b,G),_(he,G,null),l(G,$o),l(G,xe),l(G,ko),_(X,G,null),p(e,io,n),_(ue,e,n),p(e,lo,n),p(e,v,n),_(ge,v,null),l(v,Vo),l(v,ze),l(v,xo),l(v,Fe),l(v,zo),l(v,Be),l(v,Fo),l(v,$),_(fe,$,null),l($,Bo),l($,qe),l($,qo),_(S,$,null),l($,Po),l($,Pe),l($,Zo),_(D,$,null),l(v,Ao),l(v,B),_(_e,B,null),l(B,No),l(B,Ze),l(B,Go),l(B,Ae),l(B,Eo),_(Y,B,null),p(e,co,n),_(ye,e,n),p(e,po,n),p(e,P,n),_(Me,P,null),l(P,Wo),l(P,Ne),l(P,Ro),l(P,Ge),_(Te,Ge,null),p(e,mo,n),_(Ce,e,n),p(e,ho,n),p(e,V,n),_(we,V,null),l(V,Qo),l(V,Ee),l(V,Lo),l(V,We),l(V,Ho),l(V,Re),_(Ue,Re,null),p(e,uo,n),_(be,e,n),p(e,go,n),p(e,He,n),fo=!0},p(e,[n]){const E={};n&2&&(E.$$scope={dirty:n,ctx:e}),Q.$set(E);const J={};n&2&&(J.$$scope={dirty:n,ctx:e}),L.$set(J);const x={};n&2&&(x.$$scope={dirty:n,ctx:e}),H.$set(x);const W={};n&2&&(W.$$scope={dirty:n,ctx:e}),X.$set(W);const I={};n&2&&(I.$$scope={dirty:n,ctx:e}),S.$set(I);const z={};n&2&&(z.$$scope={dirty:n,ctx:e}),D.$set(z);const Z={};n&2&&(Z.$$scope={dirty:n,ctx:e}),Y.$set(Z)},i(e){fo||(y(N.$$.fragment,e),y(oe.$$.fragment,e),y(te.$$.fragment,e),y(ae.$$.fragment,e),y(Q.$$.fragment,e),y(le.$$.fragment,e),y(ce.$$.fragment,e),y(de.$$.fragment,e),y(pe.$$.fragment,e),y(me.$$.fragment,e),y(L.$$.fragment,e),y(H.$$.fragment,e),y(he.$$.fragment,e),y(X.$$.fragment,e),y(ue.$$.fragment,e),y(ge.$$.fragment,e),y(fe.$$.fragment,e),y(S.$$.fragment,e),y(D.$$.fragment,e),y(_e.$$.fragment,e),y(Y.$$.fragment,e),y(ye.$$.fragment,e),y(Me.$$.fragment,e),y(Te.$$.fragment,e),y(Ce.$$.fragment,e),y(we.$$.fragment,e),y(Ue.$$.fragment,e),y(be.$$.fragment,e),fo=!0)},o(e){M(N.$$.fragment,e),M(oe.$$.fragment,e),M(te.$$.fragment,e),M(ae.$$.fragment,e),M(Q.$$.fragment,e),M(le.$$.fragment,e),M(ce.$$.fragment,e),M(de.$$.fragment,e),M(pe.$$.fragment,e),M(me.$$.fragment,e),M(L.$$.fragment,e),M(H.$$.fragment,e),M(he.$$.fragment,e),M(X.$$.fragment,e),M(ue.$$.fragment,e),M(ge.$$.fragment,e),M(fe.$$.fragment,e),M(S.$$.fragment,e),M(D.$$.fragment,e),M(_e.$$.fragment,e),M(Y.$$.fragment,e),M(ye.$$.fragment,e),M(Me.$$.fragment,e),M(Te.$$.fragment,e),M(Ce.$$.fragment,e),M(we.$$.fragment,e),M(Ue.$$.fragment,e),M(be.$$.fragment,e),fo=!1},d(e){e&&(r(u),r(s),r(c),r(d),r(w),r(Xe),r(Se),r(R),r(De),r(Ye),r(se),r(Oe),r(ne),r(Ke),r(re),r(eo),r(oo),r(ie),r(to),r(so),r(no),r(q),r(ro),r(ao),r(b),r(io),r(lo),r(v),r(co),r(po),r(P),r(mo),r(ho),r(V),r(uo),r(go),r(He)),r(o),T(N,e),T(oe,e),T(te,e),T(ae,e),T(Q,e),T(le,e),T(ce),T(de,e),T(pe),T(me),T(L),T(H),T(he),T(X),T(ue,e),T(ge),T(fe),T(S),T(D),T(_e),T(Y),T(ye,e),T(Me),T(Te),T(Ce,e),T(we),T(Ue),T(be,e)}}}const At='{"title":"Command A Vision","local":"command-a-vision","sections":[{"title":"Overview","local":"overview","sections":[],"depth":2},{"title":"Usage tips","local":"usage-tips","sections":[],"depth":2},{"title":"Cohere2VisionConfig","local":"transformers.Cohere2VisionConfig","sections":[],"depth":2},{"title":"Cohere2VisionForConditionalGeneration","local":"transformers.Cohere2VisionForConditionalGeneration","sections":[],"depth":2},{"title":"Cohere2VisionModel","local":"transformers.Cohere2VisionModel","sections":[],"depth":2},{"title":"Cohere2VisionImageProcessor","local":"transformers.Cohere2VisionImageProcessor","sections":[],"depth":2},{"title":"Cohere2VisionProcessor","local":"transformers.Cohere2VisionProcessor","sections":[],"depth":2}],"depth":1}';function Nt(U){return wt(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class Dt extends Ut{constructor(o){super(),bt(this,o,Nt,Zt,Ct,{})}}export{Dt as component};

Xet Storage Details

Size:
80.3 kB
·
Xet hash:
4ea8304457d0c471ec1258e01a9e55dfcf420864d752b41cdfabc56fa1d06403

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