tsdat 0.2.5

A data processing framework used to convert time series data into standardized format.

Time Series Data Analytical Toolkit (TSDAT)

The Time Series Data Analytical Toolkit (TSDAT) is an open-source python framework for creating pipelines to read-in, standardize, and enhance time series datasets of any dimensionality for use in scalable applications and data repositories.

Important Links

• Tsdat Documentation: https://tsdat.readthedocs.io
• Xarray Documentation: https://xarray.pydata.org
• Template Repositories: https://github.com/tsdat/template-repositories
• Issues: https://github.com/tsdat/tsdat/issues

Getting Started

tsdat uses xarray in some capacity for nearly all components of its data pipelines, so we highly recommend checking out their documentation if you have not used xarray before or if you need a refresher.

We recommend starting by reading the docs to get a high-level overview of tsdat and the following components:

• FileHandler and Storage classes for abstracting I/O
• Pipeline and IngestPipeline base classes for creating tsdat data pipelines
• QualityChecker and QualityHandler classes for testing and managing data quality
• Pipeline Config and Storage Config yaml configuration files

After you have a basic understanding of the various tsdat components, we recommend using the local ingest template to create a data ingestion pipeline that runs on your computer. Follow the instructions outlined there to install the dependencies and run the included example.

Alternatively, if you want to start from scratch and use tsdat without starting from a template, you can install tsdat with pip like so:

pip install tsdat

and then use tsdat however you like in your project.

Contributing

We enthusiastically welcome contributions to any of our repositories.

If you find a bug or want to submit a feature request, please submit an issue. If you are submitting an issue for a bug, please explain the bug and provide a minimal example that reproduces the bug. Feature requests should clearly explain what the new feature is and what the benefit of this feature would be.

If you know how to fix a bug or implement a feature request and would like to contribute code to help resolve an open issue, please submit a pull request. See below for guidelines on how to get started with a pull request:

1. Fork tsdat to <your_username>/tsdat and clone it to your working area.
2. Install tsdat from source by running pip install -e . from the cloned repository.
3. Verify that you can still reproduce the bug or that the feature has not been implemented yet.
4. Make your changes. Be sure to test and to update the docs/ folder if appropriate.
5. Ensure your changes on your remote fork of tsdat and submit a PR.