Forecasting platforms & libraries · PyTorch Forecasting community (open source)

PyTorch Forecasting

Open-source deep-learning forecasting library built on PyTorch and PyTorch Lightning for panel/multi-series data.

Visit PyTorch Forecasting Alternatives

PyTorch Forecasting is a library for building neural forecasting models — Temporal Fusion Transformer, N-BEATS, DeepAR-style recurrent networks and others — on top of PyTorch and PyTorch Lightning, aimed at teams that want deep learning specifically rather than statistical or gradient-boosting methods. Its TimeSeriesDataSet abstraction handles the tedious parts of panel-data forecasting: encoding multiple related series, static and time-varying covariates, and variable-length histories, into batches a neural network can train on. It requires more setup and machine-learning expertise than Prophet or Nixtla's statistical models, and benefits from a GPU for training, but gives more modeling flexibility for large panels of related series (many stores, many SKUs, many sensors) where covariate interactions matter. It integrates naturally with the broader PyTorch and Lightning ecosystem for experiment tracking and deployment.

At a glance

Vendor PyTorch Forecasting community (open source)
Pricing model Open source + paid options
Free tier Yes
Deployment Self-hosted
Open source Yes (MIT)
Best for ML engineers forecasting large panels of related series (many products, stores or sensors) who want deep-learning models with covariates.

Pricing

Free and open source with no commercial tier.

Pricing has not been verified yet — see the vendor's site.

Features

  • Temporal Fusion Transformer, N-BEATS, DeepAR-style and other neural forecasters
  • TimeSeriesDataSet abstraction for panel/multi-series data
  • Static and time-varying covariate support
  • Built on PyTorch Lightning for training loops and logging
  • Attention-based interpretability for Temporal Fusion Transformer
  • Quantile/probabilistic forecasting output
  • GPU-accelerated training for large panels

Integrations

Profile last reviewed September 21, 2026

Alternatives

Darts Open-source Python library offering one unified API across dozens of classical, ML and deep-learning forecasting models. GluonTS Open-source Python toolkit from AWS for building and evaluating probabilistic deep-learning forecasting models. AutoGluon Open-source AutoML library from AWS that automates model selection and ensembling, including a time-series forecasting module.

PyTorch Forecasting in the index now

Terms to know

Time series forecasting Predicting future values of a metric using patterns found in its own historical sequence over time. Machine learning Building models that learn patterns from data to make predictions or decisions, rather than following hand-coded rules. Predictive analytics Analytics that uses historical data and statistical models to estimate what is likely to happen next. Deep learning Machine learning using neural networks with many layers, able to learn complex patterns directly from raw data such as images or text. Neural network A model made of layers of connected nodes that learns to map inputs to outputs by adjusting connection weights during training.

Related guides

How to choose a forecasting platform or library Pick a time-series forecasting library by how many series you're forecasting and how much modeling control your team wants.