JupyterLab Integration
PyPI Package: trainwave-jupyter | Latest Version: 0.1.2
Seamlessly run your Jupyter notebooks on powerful GPU infrastructure with Trainwave.ai. The Trainwave Jupyter Extension brings the power of cloud GPU computing directly into your JupyterLab environment. Transform your notebooks into scalable GPU jobs with just a few clicks, without leaving your development environment.
✨ Features
- 🚀 One-Click Job Launch: Convert notebooks to GPU jobs directly from the toolbar
- 🔐 Secure Authentication: Integrated login with your Trainwave.ai account
- 📊 Real-time Job Monitoring: Track job status and progress in real-time
- ⚙️ Flexible Configuration: Customize GPU types, counts, and project settings
- 📱 Modern UI: Clean, intuitive interface that integrates seamlessly with JupyterLab
- 🔄 Auto-save: Automatically saves your notebook before launching jobs
- 📈 Job History: View and manage your recent jobs from the extension
📦 Installation
Prerequisites
- Python 3.9 or higher
- JupyterLab 4.0 or higher
Install from PyPI
pip install trainwave-jupyter
Verify Installation
After installation, restart JupyterLab and look for the Trainwave icon in your notebook toolbar.
🚀 Quick Start
1. Sign In to Trainwave
- Open a Jupyter notebook
- Click the Trainwave icon in the toolbar
- Click “Sign In” and authenticate with your Trainwave.ai account
2. Configure Your Settings
- Click the settings icon in the Trainwave dropdown
- Select your organization and project
- Choose your preferred GPU type and count
- Save your configuration
3. Launch Your First Job
- Open or create a notebook with your code
- Click the Trainwave icon in the toolbar
- Click “Launch Job”
- Enter a name for your job
- Your notebook will be automatically saved and submitted to Trainwave
4. Monitor Your Jobs
- View active jobs in the Trainwave dropdown
- Click on job names to open them in the Trainwave web interface
- Jobs are automatically polled for status updates
📖 Detailed Usage
Job Configuration
Configure your jobs through the settings dialog:
- Organization & Project: Select your workspace and project
- GPU Type: Choose from available GPU types (CPU, T4, V100, A100, etc.)
- GPU Count: Specify the number of GPUs for your job
- Job Naming: Customize job names or use automatic naming
Job Management
- Launch Jobs: Convert any notebook to a GPU job
- Monitor Status: Real-time updates on job progress
- Access Results: Direct links to view jobs in the Trainwave web interface
- Job History: View recent jobs and their status
🔧 Configuration
Environment Variables
You can configure the extension using environment variables:
export TRAINWAVE_API_ENDPOINT="https://backend.trainwave.ai"
export TRAINWAVE_POLLING_INTERVAL=10 # seconds
export TRAINWAVE_POLLING_TIMEOUT=300 # seconds
JupyterLab Settings
Access extension settings through JupyterLab’s settings system:
- Go to Settings → Advanced Settings Editor
- Select “Trainwave Jupyter Extension”
- Modify configuration as needed
🛠️ Troubleshooting
Extension Not Appearing
If you don’t see the Trainwave icon in your notebook toolbar:
- Restart JupyterLab after installation
- Check installation:
jupyter labextension list
- Manually enable (if needed):
jupyter serverextension enable --py trainwave-jupyter jupyter labextension enable trainwave-jupyter
Authentication Issues
- Clear browser cache and try signing in again
- Check network connectivity to trainwave.ai
- Verify API endpoint in settings if using custom configuration
Job Launch Failures
- Check notebook path: Ensure your notebook is saved
- Verify settings: Confirm organization and project are selected
- Check API key: Ensure you’re properly authenticated
- Review logs: Check browser console for error messages
Performance Issues
- Reduce polling frequency in settings for better performance
- Close unused notebooks to free up resources
- Check network latency to Trainwave servers