jupyter

# Jupyter - ML/AI Development Environment

## Overview

The `jupyter` command manages JupyterLab instances for ML/AI development using Podman Quadlet containers. Each instance runs as a systemd user service with optional GPU acceleration.

**Key Concept:** Multi-instance support allows running multiple isolated JupyterLab environments simultaneously, each on different ports with different GPU configurations.

## Quick Reference

| Action | Command | Description |
|--------|---------|-------------|
| Config | `ujust jupyter config [--instance=N] [--port=...] [--gpu-type=...]` | Configure instance N |
| Start | `ujust jupyter start [--instance=N\|all]` | Start instance(s) |
| Stop | `ujust jupyter stop [--instance=N\|all]` | Stop instance(s) |
| Restart | `ujust jupyter restart [--instance=N\|all]` | Restart instance(s) |
| Logs | `ujust jupyter logs [--instance=N] [--lines=...]` | View logs |
| Status | `ujust jupyter status [--instance=N]` | Show instance status |
| URL | `ujust jupyter url [--instance=N]` | Show access URL |
| Shell | `ujust jupyter shell [--instance=N] [-- CMD...]` | Open shell in container |
| Token enable | `ujust jupyter token-enable [--instance=N]` | Enable token auth |
| Token show | `ujust jupyter token-show [--instance=N]` | Show token |
| Token disable | `ujust jupyter token-disable [--instance=N]` | Disable token auth |
| Token regenerate | `ujust jupyter token-regenerate [--instance=N]` | Generate new token |
| Delete | `ujust jupyter delete [--instance=N\|all]` | Remove instance(s) and images |

## Parameters

| Parameter | Long Flag | Short | Default | Description |
|-----------|-----------|-------|---------|-------------|
| Instance | `--instance` | `-n` | `1` | Instance number (1, 2, 3...) |
| Port | `--port` | `-p` | `8888` | Web UI port |
| GPU Type | `--gpu-type` | `-g` | `auto` | GPU type: `nvidia`, `amd`, `intel`, `none`, `auto` |
| Image | `--image` | `-i` | (default image) | Container image |
| Tag | `--tag` | `-t` | `stable`