DatacleanAir 0.1.3

Data cleaning tools for air quality analysis

DatacleanAir

A Python library for preprocessing, cleaning, and filtering air quality datasets.

Citation

Muñiz-Torres, L. F., et al. (2025). DatacleanAir: A Python library for air quality data cleaning. GitHub Repository.

Installation

pip install dataclenair

Quick Start

import pandas as pd
from DatacleanAir import kalman_filter, hampel_filter, z_score_normalize

# Load your dataset
df = pd.read_csv("pm25_data.csv")

pm = df["PM2.5"].values

# Apply normalization
pm_norm = z_score_normalize(pm)

# Apply Kalman filter
pm_kalman = kalman_filter(pm_norm)

# Remove outliers with Hampel
pm_clean = hampel_filter(pm_kalman, window_size=15, n_sigmas=3)

Features

• Z-score normalization
• Kalman filtering
• Hampel outlier detection
• Tools for common air quality preprocessing pipelines

License

MIT License