README.md

metadata

title: BERT CryptoTalk
emoji: 📈
colorFrom: gray
colorTo: yellow
sdk: gradio
sdk_version: 6.14.0
app_file: app.py
pinned: false
license: mit
short_description: BERT-CryptoTalk
hf_oauth: true
hf_oauth_scopes:
  - inference-api

🪙 CryptoBERT Gradio Inference App

A clean, modern Gradio interface for CryptoBERT, a fine-tuned BERT model for cryptocurrency-related text classification — complete with Hugging Face authentication management, dynamic token updates, and integrated sample dataset previews.

---

📦 Features

✅ CryptoBERT Text Classification
✅ Gradio UI with Sidebar & Tabs
✅ Dynamic Hugging Face Token Authentication
✅ Secure Token Management with .env Support
✅ Sample Dataset Preview (AG News)
✅ Structured, Testable Python Codebase
✅ PEP 8 & Security Best Practices
✅ Unit Testing Included

---

📸 Demo Preview

---

🚀 Quickstart

1️⃣ Clone the Repository

git clone https://github.com/your-username/crypto-bert-inference.git
cd crypto-bert-inference

2️⃣ Install Dependencies

pip install -r requirements.txt

3️⃣ Configure Environment Variables

Create a .env file:

cp .env.example .env

Edit .env:

HUGGINGFACE_TOKEN=your_huggingface_token_here

4️⃣ Launch the App

python app.py

The Gradio interface will open in your browser.

---

📊 Usage

🔐 Authenticate with Hugging Face

• Enter your Hugging Face token in the sidebar
• Click Set Token to authenticate dynamically

📝 Run Crypto Sentiment Inference

• Switch to the CryptoTalk Inference tab
• Enter cryptocurrency-related text
• Click Predict to view sentiment predictions

📖 View Sample Dataset

• The sidebar displays a preview sample from the AG News dataset for demo purposes

---

🗂️ Project Structure

crypto-bert-inference/
├── app.py
├── pipeline_utils.py
├── .env
├── requirements.txt
├── tests/
│   └── test_app.py
└── README.md

---

🧪 Run Unit Tests

python -m unittest discover tests

---

📚 Dependencies

• transformers
• gradio
• datasets
• python-dotenv
• requests

---

📖 Documentation

• CryptoBERT Model (kk08/CryptoBERT)
• Gradio Blocks API
• Hugging Face Authentication API

---

💡 Future Enhancements

• 🐳 Dockerized deployment
• 📈 Inference result charting
• 🌐 Web-hosted demo on Hugging Face Spaces

---

🛡️ Security Considerations

• Tokens are stored securely via environment variables
• No plaintext token persistence or hardcoding
• Input validation handled by Transformers pipelines
• Isolated inference processes

---

📜 License

MIT © [Canstralian(https://github.com/canstralian)

---

🖥️ Author

[Your Name] — Cybersecurity & AI Developer
GitHub | LinkedIn