# sktime/pytorch-forecasting

Time series forecasting with PyTorch

Repository: https://github.com/sktime/pytorch-forecasting
Canonical: https://ross.abutalabs.com/products/pytorch-forecasting
Homepage: https://pytorch-forecasting.readthedocs.io/
Language: Python
License: MIT
License Family: permissive
Topics: pytorch, forecasting, gpu, uncertainty, timeseries-forecasting, pytorch-lightning, deep-learning, neural-networks, timeseries, machine-learning, pandas, python, ai, data-science, temporal, artificial-intelligence, hacktoberfest
Last push: 2026-08-26T16:25:17+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 78, longevity 100
- inputs: {"age_days": 2252, "days_push": 7, "days_rel": 70, "gap_med": 79, "n_releases_24m": 10}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4977, forks 896 (observed 2026-08-28T04:09:04.104580+00:00)

## What it is
A PyTorch-based Python library for time series forecasting with state-of-the-art deep learning architectures. It provides a high-level API built on PyTorch Lightning for scaling training on GPU or CPU with automatic logging.

## Use cases
- forecast future values of time series with deep learning
- train neural forecasting models on GPU
- predict demand or sales from historical time series
- quantify uncertainty in time series forecasts
- build LSTM or Transformer-based forecasting models in Python

## When to choose
- you want a high-level API for deep learning time series forecasting in Python
- you need GPU-accelerated training with PyTorch Lightning
- you need probabilistic forecasts with uncertainty estimates
- you work with pandas DataFrames and want built-in preprocessing

## When to avoid
- you need simple statistical forecasting like ARIMA or exponential smoothing
- you want a non-PyTorch deep learning stack
- you need classical machine learning baselines only

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, data-science
- domain: machine-learning, data-science, time-series, artificial-intelligence
- platform: python, cross-platform
- tags: time-series-forecasting, pytorch, pytorch-lightning, neural-networks, uncertainty-quantification, temporal-data, gpu

## Member repositories
- sktime/pytorch-forecasting (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:04.104580+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 2026-08-29T18:17:44.145744+00:00, confidence not recorded.
  - readme: https://github.com/sktime/pytorch-forecasting (fetched 2026-08-28T04:09:04.104580+00:00, sha ea81f0375a37)
- Data as of 2026-08-30T08:39:29.467469+00:00.