metaforecast
metaforecast is a Python package for time series forecasting using meta-learning and data-centric techniques.
This package implements various techniques to improve forecasting accuracy based on dynamic model combination, data augmentation, algorithm selection, and adaptive learning, building upon Nixtla’s awesome ecosystem of state-of-the-art forecasting methods.
Features
metaforecast currently consists of five main modules:
- Dynamic Ensembles: Combining multiple models with adaptive ensemble techniques, including online learning (exponential, polynomial, and related updates), sliding-window selection, and meta-learning-based weighting (ADE).
- Synthetic Time Series Generation: Creating realistic synthetic time series data for robust model training and testing. Includes pure generators, semi-synthetic methods, transformation-based augmentation, and a callback for online data augmentation.
- Long-Horizon Meta-Learning: Instance-based meta-learning for multi-step forecasting.
- Algorithm Configuration and Selection (COSEAL): Meta-learning methods for selecting forecasting algorithms and their configurations, including MetaARIMA and ActiveTesting.
- Evaluation: Series-wise cross-validation splitters and aspect-based accuracy analysis with ModelRadar (horizon, groups, anomalies, hard series, and ROPE comparisons).
Installation
You can install metaforecast using pip:
pip install metaforecast
[Optional] Installation from source
To install metaforecast from source, clone the repository and run the following command:
git clone https://github.com/vcerqueira/metaforecast
cd metaforecast
pip install -e .
Documentation
Check the documentation for the API reference and module descriptions. You can get started with a few tutorials.
⚠️ WARNING
metaforecast is in the early stages of development. The codebase may undergo significant changes. If you encounter any issues, please report them in GitHub Issues
License
metaforecast is dual-licensed.
- AGPL-3.0-or-later for open-source use. You may use, modify, and share the source. If you distribute a modified version or run one in production, you must make the corresponding source available.
See LICENSE for the full terms.
1 maintainer
Metadata
1 maintainer
Release files for metaforecast 0.3.6
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| metaforecast-0.3.6.tar.gz | 15.2 MB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| metaforecast-0.3.6-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 27.7 MB
Release files / metaforecast-0.3.6.tar.gz
| Download URL | metaforecast-0.3.6.tar.gz |
|---|---|
| Size | 15.2 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
307e21f4b04dd7dc163ec0691ab042df1f770b899536474ae64735c715654397
|
|
BLAKE2b-256 checksum How to use checksums |
f9b693c86a8e92ab6d13187683741a5ba82a719a08abdad788876f13d206691b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.13.13
|
Release files / metaforecast-0.3.6-py3-none-any.whl
| Download URL | metaforecast-0.3.6-py3-none-any.whl |
|---|---|
| Size | 12.5 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
ad8b40a16ba9ab784928bf45c267fb03e9aa2c1f51d537caf40d6deca3e8cd0d
|
|
BLAKE2b-256 checksum How to use checksums |
e4c4480260dd69e3e31414d211e6205daf93d07806c4f68b28463f5c552ae9e4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.13.13
|
Security for metaforecast 0.3.6
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
