Skip to main content

Unofficial package for PNADC microdata (IBGE)

Project description

py-pnadc: Unofficial Python Package for PNADcIBGE

🇺🇸 English / 🇧🇷 Português

:us: English

About this package

py-pnadc is a Python package containing tools for downloading and reading the microdata from the Continuous National Household Sample Survey (PNAD Contínua) from Brazilian Institute of Geography and Statistics (IBGE).

This package solves common issues related to handling large PNADC files by streaming data in stream and converting it directly to .parquet.

It provides a pythonic alternative to the R official package.

Main funcionalities

  • Analytics-ready: Converts the Fixed-Width-Files (FWF) files direct to .parquet.
  • Low RAM consumption: Processes data in chunks, enabling execution in regular notebooks.
  • Automatic ETL: Manages downloading, layout parsing and convertion in a single command.

Installation / Instalação

PyPI:

pip install py-pnadc

Developer mode:

git clone https://github.com/andre-kmp/py-pnadc.git
cd py-pnadc
pip install .

How to use it

Running the main function

import pnadc

# This will:
# 1. Download the layout and the microdata (if they don't exist yet)
# 2. Convert to .parquet
# 3. Return a pandas dataframe with the selected columns
df = pnadc.load_microdata()

print(df.head())

Setting parameters

You may override the package defaults by passing arguments directly to the load_microdata function.

  • columns_to_keep: List of columns (IBGE codes) to keep.
  • load_years: List of years to process (e.g.: ['2023', '2024']).
  • chunk_size: Number of rows per batch in memory (reduce it for low RAM environments). Default is 25,000.

Usage example:

import pnadc

# 1. Setting preferences
my_columns = ['UF', 'V2007', 'V2009', 'VD4001', 'VD4002']
target_years = ['2023', '2024']

# 2. Loading data with custom arguments
df = pnadc.load_microdata(
    columns_to_keep=my_columns,
    load_years=target_years,
    chunk_size=10000
)

print(f"Loaded data: {df.shape}")
print(df.head())

:warning: Disclaimer

This is an unofficial project and is not affiliated with IBGE.

:brazil: Português

Sobre este pacote

py-pnadc é um pacote Python com ferramentas para baixar e ler os microdados da Pesquisa Nacional de Domicílios Contínua (PNADC) do Instituto Brasileiro de Geografia e Estatística (IBGE).

Esse pacote resolve os problemas comuns ao lidar com os arquivos de texto gigantes da PNADC, processando os dados em stream e convertendo-os para .parquet.

É uma alternativa pythonica ao pacote oficial em R.

Principais funcionalidades

  • Eficiente para análises: Converte os arquivos de largura fixa (FWF) diretamente para .parquet.
  • Baixo consumo de RAM: Processa os dados em chunks, o que permite rodar em notebooks comuns.
  • ETL automático: Gerencia o download, leitura do layout e conversão em um único comando.

Instalação

Via PyPI:

pip install py-pnadc

Modo de desenvolvedor:

git clone https://github.com/andre-kmp/py-pnadc.git
cd py-pnadc
pip install .

Como usar

Rodando a função principal

import pnadc

# Isso irá:
# 1. Baixar o dicionário e os dados (se ainda não existirem)
# 2. Converter para Parquet
# 3. Retornar um pandas DataFrame com as colunas selecionadas
df = pnadc.load_microdata()

print(df.head())

Configurando os parâmetros

Você pode sobrescrever os padrões do pacote passando argumentos diretamente para a função load_microdata:

  • columns_to_keep: Lista de colunas (códigos do IBGE) para manter.
  • load_years: Lista de anos a serem processados (ex: ['2023', '2024']).
  • chunk_size: Número de linhas por lote na memória (diminua se tiver pouca RAM). Por default, 25000.

Exemplo de uso:

import pnadc

# 1. Definindo suas preferências
minhas_colunas = ['UF', 'V2007', 'V2009', 'VD4001', 'VD4002']
anos_alvo = ['2023', '2024']

# 2. Carregando os dados com os seus argumentos
df = pnadc.load_microdata(
    columns_to_keep=minhas_colunas,
    load_years=anos_alvo,
    chunk_size=10000
)

print(f"Dados carregados: {df.shape}")
print(df.head())

:warning: Isenção de responsabilidade

Este é um projeto não-oficial e não possui afiliação com o IBGE.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

py_pnadc-0.1.5.tar.gz (10.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

py_pnadc-0.1.5-py3-none-any.whl (9.9 kB view details)

Uploaded Python 3

File details

Details for the file py_pnadc-0.1.5.tar.gz.

File metadata

  • Download URL: py_pnadc-0.1.5.tar.gz
  • Upload date:
  • Size: 10.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for py_pnadc-0.1.5.tar.gz
Algorithm Hash digest
SHA256 3267e866cd7cc6b38b1157ffd5e98b56f6563ecd115c3a2ac35848340f1a441b
MD5 b121d050823d98e154222dc7107fe872
BLAKE2b-256 6e310eca5de0bafefed7921cde91bdd11d88f92bec159c237b187a7946197ac6

See more details on using hashes here.

Provenance

The following attestation bundles were made for py_pnadc-0.1.5.tar.gz:

Publisher: python-publish.yml on andre-kmp/py-pnadc

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file py_pnadc-0.1.5-py3-none-any.whl.

File metadata

  • Download URL: py_pnadc-0.1.5-py3-none-any.whl
  • Upload date:
  • Size: 9.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for py_pnadc-0.1.5-py3-none-any.whl
Algorithm Hash digest
SHA256 5f2c06d75341aec9ca8f3463a119ff3791c1d3cf0cd66b72bcc3b59094ddac7f
MD5 94d273d803775f2fb7b0401b4e1b4950
BLAKE2b-256 e3c61500ec0da9f604735ed6883cfaa58f5083ec60d007f4a59da52353eec3de

See more details on using hashes here.

Provenance

The following attestation bundles were made for py_pnadc-0.1.5-py3-none-any.whl:

Publisher: python-publish.yml on andre-kmp/py-pnadc

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page