timedatamodel

A lightweight data model for time series data with pandas, numpy, polars, and xarray support.

Features

• TimeSeries – univariate time series with rich metadata (unit, frequency, location, …)

• TimeSeriesTable – multivariate time series sharing the same index

• TimeSeriesArray – N-dimensional time series with named dimensions and label-based selection

• TimeSeriesCollection – heterogeneous container for series that don’t share an index

• Built-in conversions to/from pandas, numpy, and polars

• Enum-based frequency (ISO 8601 durations) and data-type annotations

• Hierarchical time series – define aggregation trees and reconcile forecasts

• Optional pint unit support and shapely geo-location metadata

Getting Started

• Installation
• Overview
• Basic Usage

Tutorials

• Getting Started with TimeDataModel
• NumPy and Pandas Transforms
• Unit Handling and Validation
• TimeSeriesList Operations
• Multivariate Time Series
• Arrays and Collections
• Data Quality and Coverage
• I/O and Interoperability
• Geographical Support
• Hierarchical Time Series

API Reference

• API Reference