# awslabs/gluonts

Probabilistic time series modeling in Python

Repository: https://github.com/awslabs/gluonts
Canonical: https://ross.abutalabs.com/products/gluonts
Homepage: https://ts.gluon.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: time-series, deep-learning, forecasting, neural-networks, machine-learning, time-series-prediction, time-series-forecasting, mxnet, pytorch, aws, sagemaker, timeseries, artificial-intelligence, data-science, torch
Last push: 2026-07-31T11:26:21+00:00

## Health v2 (maintenance only)
Score: 88/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 95, release rhythm 71, longevity 100
- inputs: {"age_days": 2667, "days_push": 33, "days_rel": 33, "gap_med": 113.5, "n_releases_24m": 5}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5227, forks 830 (observed 2026-08-28T04:09:13.283376+00:00)

## What it is
GluonTS is a Python library for probabilistic time series modeling, focused on deep learning based forecasting models built on PyTorch. It provides dataset handling, model training, and evaluation utilities for time series prediction tasks.

## Use cases
- forecast future values of a time series with uncertainty estimates
- train a DeepAR model on my sales data
- probabilistic demand forecasting in Python
- evaluate and compare time series forecasting models
- train deep learning models on multiple related time series
- generate probabilistic predictions from pandas DataFrames

## When to choose
- you need probabilistic (distributional) forecasts rather than point predictions
- you want ready-to-use deep learning forecasting models like DeepAR with PyTorch
- you need tooling for datasets, splitting, and evaluation of time series experiments

## When to avoid
- you only need simple classical methods like ARIMA or exponential smoothing
- you need a lightweight point-forecast library without deep learning dependencies
- you need zero-shot forecasting with pretrained foundation models (consider Chronos instead)

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, deep-learning, data-science
- domain: machine-learning, data-science, time-series, artificial-intelligence
- platform: python
- tags: time-series, forecasting, probabilistic-forecasting, pytorch, deepar, neural-networks

## Member repositories
- awslabs/gluonts (main) score 88

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:13.283376+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:59:40.724786+00:00, confidence not recorded.
  - readme: https://github.com/awslabs/gluonts (fetched 2026-08-28T04:09:13.283376+00:00, sha 91a46c5b6019)
  - homepage: https://ts.gluon.ai (fetched 2026-08-29T08:55:07.587682+00:00, sha 8cea8dade42d)
  - site_page: https://ts.gluon.ai/stable/getting_started/install.html (fetched 2026-08-29T08:55:07.590517+00:00, sha 016ecf43825f)
  - registry_pypi: https://pypi.org/pypi/gluonts/json (fetched 2026-08-29T08:55:07.595590+00:00, sha c42a8f19187c)
  - site_page: https://ts.gluon.ai/stable/api/gluonts/gluonts.html (fetched 2026-08-29T08:55:07.592445+00:00, sha 5b1cce86fe85)
- Data as of 2026-08-30T08:39:29.467469+00:00.