JupyterLab¶

Deep dive into the JupyterLab configuration and features.

Overview¶

JupyterLab is the next-generation interface for Jupyter notebooks. In DataSci Homelab, it's configured with:

• Python and R kernels
• Git integration
• Notebook diffing
• Interactive widgets
• Persistent package storage

Accessing JupyterLab¶

URL: http://localhost:8888

Authentication: Enter the token from your JUPYTER_TOKEN in .env

Available Kernels¶

Python 3¶

The default Python kernel with the full scientific stack.

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

# Your Python code here

R¶

Use R directly in Jupyter via IRkernel.

library(tidyverse)

ggplot(mtcars, aes(mpg, hp)) +
  geom_point()

Package Management¶

Python Packages¶

# Install in a notebook cell
!pip install packagename

# Or with specific version
!pip install packagename==1.2.3

Packages are installed to ~/.local/lib/python3.x/site-packages/, which is volume-mounted for persistence.

R Packages (in R Kernel)¶

install.packages("packagename")

Uses the same R library as RStudio Server.

Verifying Installation¶

# Python
import packagename
print(packagename.__version__)

# R
library(packagename)
packageVersion("packagename")

Extensions¶

Pre-installed Extensions¶

Git Extension¶

Access via the Git icon in the left sidebar:

1. Clone repositories
2. Stage/unstage changes
3. Commit with messages
4. Push/pull to remotes
5. View history and diffs

Using Widgets¶

import ipywidgets as widgets

slider = widgets.IntSlider(
    value=50,
    min=0,
    max=100,
    description='Value:'
)
display(slider)

Working with Files¶

File Browser¶

The left sidebar file browser shows:

• /home/rstudio/ - Your home directory
• /data/ - Shared data directory

Uploading Files¶

1. Click the upload button in the file browser
2. Or drag-and-drop files directly

Downloading Files¶

Right-click on any file → Download

Data Science Workflows¶

Loading Data¶

import pandas as pd

# From CSV
df = pd.read_csv('/data/myfile.csv')

# From shared directory
df = pd.read_csv('/data/shared/dataset.csv')

# From Parquet (fast!)
df = pd.read_parquet('/data/large_dataset.parquet')

Visualization¶

import matplotlib.pyplot as plt

plt.figure(figsize=(10, 6))
plt.plot(df['x'], df['y'])
plt.xlabel('X axis')
plt.ylabel('Y axis')
plt.title('My Plot')
plt.show()

import seaborn as sns

sns.set_theme()
sns.scatterplot(data=df, x='x', y='y', hue='category')

import plotly.express as px

fig = px.scatter(df, x='x', y='y', color='category')
fig.show()

Database Connections¶

from sqlalchemy import create_engine
import pandas as pd

# PostgreSQL
engine = create_engine('postgresql://user:pass@host:5432/dbname')

# Query to DataFrame
df = pd.read_sql("SELECT * FROM table", engine)

# DuckDB (embedded)
import duckdb
con = duckdb.connect('mydata.duckdb')
df = con.execute("SELECT * FROM table").df()

R-Python Integration¶

Use rpy2 to mix R and Python in the same notebook:

%load_ext rpy2.ipython

%%R
library(ggplot2)
ggplot(mtcars, aes(mpg, hp)) + geom_point()

# Pass data from Python to R
%%R -i df
summary(df)

Quarto in Jupyter¶

Create Quarto documents from Jupyter notebooks:

Method 1: Convert Notebook¶

quarto render notebook.ipynb --to html
quarto render notebook.ipynb --to pdf

Method 2: Native Quarto¶

Create .qmd files and open them in JupyterLab's text editor, then render:

quarto render document.qmd

Keyboard Shortcuts¶

Command Mode (Blue border)¶

Edit Mode (Green border)¶

Configuration¶

JupyterLab Settings¶

Access via Settings → Settings Editor

Common customizations:

• Theme (dark/light)
• Font size
• Auto-save interval
• Keyboard shortcuts

Server Configuration¶

The server is configured via /etc/jupyter/jupyter_server_config.py:

c.ServerApp.ip = '0.0.0.0'
c.ServerApp.port = 8888
c.ServerApp.open_browser = False
c.ServerApp.allow_origin = '*'

Tips & Tricks¶

Magic Commands¶

# Time a cell
%%time
expensive_operation()

# Time multiple runs
%%timeit
operation()

# Run shell commands
!ls -la

# Environment variables
%env MY_VAR=value

# Load external scripts
%run script.py

Rich Output¶

# Display multiple outputs
from IPython.display import display

display(df1)
display(df2)

Autoreload¶

%load_ext autoreload
%autoreload 2

# Now modules reload automatically when changed
import mymodule

Troubleshooting¶

Kernel Not Starting¶

# Check kernel list
docker-compose exec homelab jupyter kernelspec list

# Reinstall Python kernel
docker-compose exec homelab python -m ipykernel install --user

# Reinstall R kernel
docker-compose exec homelab Rscript -e "IRkernel::installspec()"

Token Not Working¶

# Check the configured token
docker-compose exec homelab jupyter server list

# Or check logs
docker-compose logs homelab | grep token

Extension Not Working¶

# List extensions
docker-compose exec homelab jupyter labextension list

# Rebuild JupyterLab
docker-compose exec homelab jupyter lab build

Notebook Won't Save¶

Check permissions:

docker-compose exec homelab ls -la /home/rstudio/

If permissions are wrong:

docker-compose exec homelab chown -R rstudio:rstudio /home/rstudio/