Spaces:
Sleeping
Sleeping
Upload components/Datasets.jsx with huggingface_hub
Browse files- components/Datasets.jsx +74 -0
components/Datasets.jsx
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import React from 'react';
|
| 2 |
+
|
| 3 |
+
const Datasets = () => {
|
| 4 |
+
const datasets = [
|
| 5 |
+
{
|
| 6 |
+
name: "Web Text Corpus",
|
| 7 |
+
size: "2.5TB",
|
| 8 |
+
languages: "50+",
|
| 9 |
+
description: "Diverse web content from reputable sources across multiple domains"
|
| 10 |
+
},
|
| 11 |
+
{
|
| 12 |
+
name: "Academic Papers",
|
| 13 |
+
size: "800GB",
|
| 14 |
+
languages: "English",
|
| 15 |
+
description: "Scientific papers and research documents from top conferences"
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
name: "Code Repository",
|
| 19 |
+
size: "1.2TB",
|
| 20 |
+
languages: "10+",
|
| 21 |
+
description: "Open source code across multiple programming languages"
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
name: "Multilingual News",
|
| 25 |
+
size: "1.8TB",
|
| 26 |
+
languages: "25+",
|
| 27 |
+
description: "Global news articles with temporal metadata"
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
name: "Conversational Data",
|
| 31 |
+
size: "500GB",
|
| 32 |
+
languages: "15+",
|
| 33 |
+
description: "Dialogues and conversational exchanges from various platforms"
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
name: "Technical Documentation",
|
| 37 |
+
size: "300GB",
|
| 38 |
+
languages: "8+",
|
| 39 |
+
description: "API docs, manuals, and technical guides"
|
| 40 |
+
}
|
| 41 |
+
];
|
| 42 |
+
|
| 43 |
+
return (
|
| 44 |
+
<section id="datasets" className="py-20 bg-gray-50">
|
| 45 |
+
<div className="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8">
|
| 46 |
+
<div className="text-center mb-16">
|
| 47 |
+
<h2 className="text-3xl md:text-4xl font-bold text-gray-900 mb-4">
|
| 48 |
+
Featured Datasets
|
| 49 |
+
</h2>
|
| 50 |
+
<p className="text-xl text-gray-600 max-w-2xl mx-auto">
|
| 51 |
+
Comprehensive collections tailored for different LLM training scenarios
|
| 52 |
+
</p>
|
| 53 |
+
</div>
|
| 54 |
+
<div className="grid grid-cols-1 md:grid-cols-2 lg:grid-cols-3 gap-6">
|
| 55 |
+
{datasets.map((dataset, index) => (
|
| 56 |
+
<div key={index} className="bg-white rounded-xl shadow-sm border p-6 hover:shadow-md transition-shadow">
|
| 57 |
+
<h3 className="text-xl font-semibold text-gray-900 mb-2">{dataset.name}</h3>
|
| 58 |
+
<div className="flex gap-4 text-sm text-gray-600 mb-3">
|
| 59 |
+
<span className="bg-primary-50 text-primary-700 px-2 py-1 rounded">Size: {dataset.size}</span>
|
| 60 |
+
<span className="bg-green-50 text-green-700 px-2 py-1 rounded">{dataset.languages} languages</span>
|
| 61 |
+
</div>
|
| 62 |
+
<p className="text-gray-600">{dataset.description}</p>
|
| 63 |
+
<button className="mt-4 text-primary-600 hover:text-primary-700 font-semibold transition-colors">
|
| 64 |
+
Learn More →
|
| 65 |
+
</button>
|
| 66 |
+
</div>
|
| 67 |
+
))}
|
| 68 |
+
</div>
|
| 69 |
+
</div>
|
| 70 |
+
</section>
|
| 71 |
+
);
|
| 72 |
+
};
|
| 73 |
+
|
| 74 |
+
export default Datasets;
|