notebookux is a small visual layer for Python notebooks.
It provides reusable presentation helpers for academic and technical notebooks:
- themes
- cover and section pages
- badges, cards, grids and code panels
- styled tables and figures
- simple multi-screen modules with previous/next navigation
The package is UI-only. It does not depend on Colab and does not implement data processing, machine learning pipelines or project-specific logic.
Documentation is published with MkDocs Material:
- GitHub Pages: https://jyrikes.github.io/notebookux/
- API reference: https://jyrikes.github.io/notebookux/api/
cd "C:\Users\Yngrid Kalinne\Documents\cq-novo\notebookux"
python -m pip install -e ".[dev,docs,examples]"from notebookux import UX
UX.use_theme("blueprint")
UX.cover(
title="Resolucao de Computacao Quantica",
subtitle="Deutsch-Jozsa e Bernstein-Vazirani",
meta=[
("Disciplina", "Computacao Quantica"),
("Formato", "Notebook"),
],
bullets=[
"Execucoes com evidencias",
"Tabelas de contagens",
"Comparacao entre simuladores",
],
)Para notebooks academicos, voce pode importar com um nome curto e usar os componentes prontos sem escrever HTML manualmente.
from notebookux import UX as CQ
CQ.setup("blueprint", compact=True)
CQ.section(
"Diagnostico inicial",
subtitle="Resumo visual da base antes da modelagem.",
kicker="Etapa 1",
)
CQ.badges([
("Dataset", "blue"),
("Classificacao", "green"),
("Reprodutivel", "amber"),
])
CQ.metrics([
("Linhas", "53.940", "Registros carregados"),
{"label": "Classes", "value": "5", "note": "Variavel alvo", "color": "purple"},
])
CQ.checklist([
("Dataset carregado", True),
{"label": "Modelos configurados", "done": False, "note": "Executar na proxima celula"},
])
CQ.steps([
("Carregamento", "Obter a base e conferir as colunas.", "done"),
("Pre-processamento", "Preparar dados para os modelos.", "active"),
("Resultados", "Comparar metricas e discutir.", "pending"),
])
CQ.compare(
"Antes",
"Dados brutos e sem organizacao visual.",
"Depois",
"Notebook com secoes, metricas e fluxo documentado.",
title="Comparacao do notebook",
)Every main visual component has a method that returns HTML and a matching method that displays it in a notebook.
card = UX.card_html("Dataset", "Resumo da base carregada.", variant="info")
table = UX.table_html(df.head(), title="Amostra")
UX.module(
"Diagnostico",
[
UX.screen("Contexto", card),
UX.screen("Tabela", table),
],
)Use from notebookux import UX as CQ when migrating notebooks that already use a
short CQ helper name.
python -m ruff check .
python -m pyright
python -m pytest --cov=notebookux --cov-report=term-missing
python -m build
python -m mkdocs build --strictReleases are published by GitHub Actions when a v* tag is pushed. The PyPI
project must be configured with Trusted Publishing for repository
jyrikes/notebookux, workflow release.yml and environment pypi.