data-cleaner

What problem does it solve?

Data quality is foundational for effective modeling; messy CSV/Excel inputs cause biased results and wasted time. The Data-Cleaner standardizes loading, inspecting, and cleaning datasets to ensure reproducible inputs for analytics and modeling.

Core Features & Use Cases

• Load and inspect data from common formats (CSV, Excel, JSON) and show shape, dtypes, missing values, and basic statistics.
• Handle missing values with robust strategies (auto, median, forward-fill, interpolate) and drop heavily incomplete columns.
• Detect and cap or remove outliers using IQR or Z-score methods.
• Normalize numeric features with standard, min-max, or robust scaling.
• Fix data types (dates to datetime, categoricals) to improve downstream analysis.
• Produce a cleaned dataset (processed.csv) and a cleaning report (where the pipeline decisions are documented).

Quick Start

To start, run the cleaning pipeline on your raw data file: clean_data(filepath='data/raw_data.csv', output_path='data/processed.csv', normalize=False). Then review the generated processed.csv and processed_report.json to confirm improvements.