projects

guneysus.github.io-archieved

โญ Canonical โ€” this is the single source of truth for this project.

Overview

Field Value
Name guneysus.github.io-archieved
Tier ๐Ÿ“š Learning & Experiments
Description - exif(exif/)
Visibility ๐Ÿ”’ Private
Fork No
Last activity ย 
Status โœ… Archived
Original source Github
Original owner guneysus
Language HTML
Type Library (Node.js)
Size Small
Idea Archived personal GitHub Pages site with blog content
Idea score 50
Impl score 60
Rating 54

๐Ÿ’ก Core Idea

๐Ÿ” Code Snippet

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Interactive Report: The Power of BERT & CodeBERT</title>
    <script src="https://cdn.tailwindcss.com"></script>
    <script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
    <link rel="preconnect" href="https://fonts.googleapis.com">
    <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
    <link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700&display=swap" rel="stylesheet">
    <!-- Chosen Palette: Warm Neutrals with Subtle Accents -->
    <!-- Application Structure Plan: The SPA is designed with a top-level tabbed navigation to switch between two main contexts: BERT (for NLP) and CodeBERT (for Software Engineering). This thematic split allows users to focus on their area of interest. Within each tab, a consistent structure is used: an 'Introduction' to the model, a 'How It Works' section with an interactive diagram, an 'Applications' section with filterable cards, and new 'LLM-Powered Features' sections. This task-oriented design was chosen to make the dense report content digestible and explorable. Users can quickly grasp the core concepts and then dive into specific use cases that are relevant to them, promoting understanding and engagement over passive reading. -->
    <!-- Visualization & Content Choices:
        - Report Info: BERT/CodeBERT architecture. Goal: Explain complex concepts. Viz: Interactive HTML/CSS diagram. Interaction: Hover to highlight components (e.g., Encoder, Attention). Justification: A static diagram would be dense. Interaction breaks down the complexity. Library: Vanilla JS + Tailwind.
        - Report Info: Application lists for BERT/CodeBERT. Goal: Organize and compare uses. Viz: Filterable card-based grid. Interaction: Buttons to filter by category (e.g., 'Search', 'Code Analysis'). Justification: Allows users to quickly find relevant applications instead of reading a long list. Library: Vanilla JS.
        - Report Info: Domain-specific BERT models. Goal: Compare variety. Viz: Bar chart. Interaction: Tooltips on hover provide details. Justification: Visually represents the proliferation and focus of specialized models. Library: Chart.js.
        - New Feature: Text Summarization (BERT tab). Goal: Demonstrate BERT's NLP capabilities. Viz: Text input/output. Interaction: Button to trigger LLM call. Justification: Directly showcases a core NLP use case. Library: Gemini API (gemini-2.5-flash-preview-05-20).
        - New Feature: Code Explanation (CodeBERT tab). Goal: Demonstrate CodeBERT's bimodal understanding. Viz: Code input/text output. Interaction: Button to trigger LLM call. Justification: Highlights the unique ability to bridge natural and programming languages. Library: Gemini API (gemini-2.5-flash-preview-05-20).
        - All content is presented in concise, summarized blocks to fit the interactive, dashboard-like feel of the application.
    -->
    <!-- CONFIRMATION: NO SVG graphics used. NO Mermaid JS used. -->
    <style>
        body {
            font-family: 'Inter', sans-serif;

Archive Origins


โ† chess.js openresty-twitter-clone โ†’

Generated by scripts/generate.py ยท Back to index