|
<!DOCTYPE html> |
|
<html> |
|
<head> |
|
<meta charset="UTF-8" /> |
|
<meta name="viewport" content="width=device-width, initial-scale=1.0" /> |
|
<script src="https://cdn.tailwindcss.com"></script> |
|
|
|
<script src="https://unpkg.com/[email protected]/dist/es-module-shims.js"></script> |
|
<script type="importmap"> |
|
{ |
|
"imports": { |
|
"@huggingface/inference": "https://cdn.jsdelivr.net/npm/@huggingface/[email protected]/+esm" |
|
} |
|
} |
|
</script> |
|
</head> |
|
<body> |
|
<form class="w-[90%] mx-auto pt-8" onsubmit="launch(); return false;"> |
|
<h1 class="text-3xl font-bold"> |
|
<span |
|
class="bg-clip-text text-transparent bg-gradient-to-r from-pink-500 to-violet-500" |
|
> |
|
Image to text demo with |
|
<a href="https://github.com/huggingface/huggingface.js"> |
|
<kbd>@huggingface/inference</kbd> |
|
</a> |
|
</span> |
|
</h1> |
|
|
|
<p class="mt-8"> |
|
First, input your token if you have one! Otherwise, you may encounter |
|
rate limiting. You can create a token for free at |
|
<a |
|
target="_blank" |
|
href="https://huggingface.co/settings/tokens" |
|
class="underline text-blue-500" |
|
>hf.co/settings/tokens</a |
|
> |
|
</p> |
|
|
|
<input |
|
type="text" |
|
id="token" |
|
class="rounded border-2 border-blue-500 shadow-md px-3 py-2 w-96 mt-6" |
|
placeholder="token (optional)" |
|
/> |
|
|
|
<p class="mt-8"> |
|
Pick the model you want to run. Check out over 100 models for image to text |
|
<a |
|
href="https://huggingface.co/tasks/image-to-text" |
|
class="underline text-blue-500" |
|
target="_blank" |
|
> |
|
here</a |
|
>. The default model is for image captioning, but you can do text extraction, ... |
|
</p> |
|
|
|
|
|
<input |
|
type="text" |
|
id="model" |
|
class="rounded border-2 border-blue-500 shadow-md px-3 py-2 w-96 mt-6" |
|
value="nlpconnect/vit-gpt2-image-captioning" |
|
required |
|
/> |
|
|
|
<p class="mt-8">Finally the input image</p> |
|
|
|
<input type="file" required accept="image/*" |
|
class="rounded border-blue-500 shadow-md px-3 py-2 w-96 mt-6 block" |
|
rows="5" |
|
id="prompt" |
|
/> |
|
|
|
<button |
|
id="submit" |
|
class="my-8 bg-green-500 rounded py-3 px-5 text-white shadow-md disabled:bg-slate-300" |
|
> |
|
Run |
|
</button> |
|
|
|
<p class="text-gray-400 text-sm">Output logs</p> |
|
<div id="logs" class="bg-gray-100 rounded p-3 mb-8 text-sm"> |
|
Output will be here |
|
</div> |
|
|
|
<p>Check out the <a class="underline text-blue-500" href="https://huggingface.co/spaces/huggingfacejs/image-to-text/blob/main/index.html" target="_blank">source code</a></p> |
|
</form> |
|
|
|
<script type="module"> |
|
import { HfInference } from "@huggingface/inference"; |
|
let running = false; |
|
async function launch() { |
|
if (running) { |
|
return; |
|
} |
|
running = true; |
|
try { |
|
const hf = new HfInference( |
|
document.getElementById("token").value.trim() || undefined |
|
); |
|
const model = document.getElementById("model").value.trim(); |
|
const prompt = document.getElementById("prompt").files[0]; |
|
document.getElementById("logs").textContent = ""; |
|
const {generated_text} = await hf.imageToText({model, data: prompt}); |
|
document.getElementById("logs").textContent = generated_text; |
|
} catch (err) { |
|
alert("Error: " + err.message); |
|
} finally { |
|
running = false; |
|
} |
|
} |
|
window.launch = launch; |
|
</script> |
|
</body> |
|
</html> |
