README.md

metadata

title: Textilindo AI Assistant
emoji: 🤖
colorFrom: blue
colorTo: green
sdk: docker
sdk_version: 4.0.0
app_file: app.py
pinned: false
license: mit
short_description: AI Assistant for Textilindo textile company

🤖 Textilindo AI Assistant

An intelligent AI assistant for Textilindo textile company with advanced training capabilities, built with FastAPI and Hugging Face Transformers.

✨ Features

• Intelligent Chat Interface: Natural language conversations in Indonesian
• Company Knowledge: Trained on Textilindo's specific information
• Model Training: Train custom models with your data
• Fast Response: Optimized for quick customer service
• Mobile Friendly: Responsive web interface
• API Ready: RESTful API for integration

🚀 Quick Start

Chat Interface

Visit the main page to start chatting with the AI assistant. Ask questions about:

• Company location and hours
• Product information
• Ordering and shipping
• Sample requests
• Pricing and terms

Training API

Start Training

curl -X POST "https://harismlnaslm-Textilindo-AI.hf.space/api/train/start" \
  -H "Content-Type: application/json" \
  -d '{
    "model_name": "distilgpt2",
    "dataset_path": "data/lora_dataset_20250910_145055.jsonl",
    "config_path": "configs/training_config.yaml",
    "max_samples": 10,
    "epochs": 1,
    "batch_size": 1,
    "learning_rate": 5e-5
  }'

Check Training Status

curl "https://harismlnaslm-Textilindo-AI.hf.space/api/train/status"

Test Trained Model

curl -X POST "https://harismlnaslm-Textilindo-AI.hf.space/api/train/test"

Get Training Data Info

curl "https://harismlnaslm-Textilindo-AI.hf.space/api/train/data"

Check GPU Availability

curl "https://harismlnaslm-Textilindo-AI.hf.space/api/train/gpu"

🛠️ Technical Details

Architecture

• Framework: FastAPI with Uvicorn
• AI Model: Llama 3.1 8B Instruct (via Hugging Face)
• Training: PyTorch with Transformers
• Language: Indonesian (Bahasa Indonesia)
• Deployment: Docker on Hugging Face Spaces

API Endpoints

Chat Endpoints

• GET / - Main chat interface
• POST /chat - Chat API endpoint
• GET /health - Health check
• GET /info - Application information

Training Endpoints

• POST /api/train/start - Start model training
• GET /api/train/status - Check training progress
• GET /api/train/data - Get training data information
• GET /api/train/gpu - Check GPU availability
• POST /api/train/test - Test trained model

Environment Variables

Set these in your space settings:

# Required: Hugging Face API Key
HUGGINGFACE_API_KEY=your_api_key_here

# Optional: Model selection
DEFAULT_MODEL=meta-llama/Llama-3.1-8B-Instruct

📞 Support

For technical issues:

1. Check the /health endpoint
2. Review space logs
3. Verify environment variables
4. Test with mock responses

---

Built with ❤️ for Textilindo customers