# sktime/pytorch-forecasting

**Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/sktime-pytorch-forecasting).**

4,787 stars · 783 forks · Python · mit

## Links

- GitHub: https://github.com/sktime/pytorch-forecasting
- Homepage: https://pytorch-forecasting.readthedocs.io/
- awesome-repositories: https://awesome-repositories.com/repository/sktime-pytorch-forecasting.md

## Topics

`ai` `artificial-intelligence` `data-science` `deep-learning` `forecasting` `gpu` `hacktoberfest` `machine-learning` `neural-networks` `pandas` `python` `pytorch` `pytorch-lightning` `temporal` `timeseries` `timeseries-forecasting` `uncertainty`

## Description

PyTorch Forecasting is a deep learning framework designed for building and training neural network architectures specifically for time series forecasting. It serves as a comprehensive toolkit for implementing autoregressive models, multi-horizon forecasting, and probabilistic prediction intervals using PyTorch tensors.

The library distinguishes itself through a probabilistic forecasting toolkit that generates prediction intervals and quantile forecasts using both parametric and non-parametric distributions. It further provides a neural network model optimizer for automated hyperparameter tuning and pruning to improve architecture efficiency.

The framework covers a broad surface of capabilities, including multivariate time series analysis and the integration of static and time-varying covariates. It includes a dedicated data pipeline for transforming tabular data into normalized tensors, as well as tools for interpretable model evaluation via variable dependency analysis and multi-horizon accuracy metrics.

The system supports distributed training across CPUs and multiple GPUs to accelerate model convergence.

## Tags

### Artificial Intelligence & ML

- [Deep Learning Forecasting](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting/deep-learning-forecasting.md) — Provides a comprehensive framework for applying deep neural networks to univariate and multivariate time series prediction.
- [Autoregressive Models](https://awesome-repositories.com/f/artificial-intelligence-ml/autoregressive-models.md) — Implements neural network architectures that generate long-term forecasts by iteratively feeding previous outputs back as inputs.
- [Probabilistic](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/architectures/neural-network-components/loss-functions/probabilistic.md) — Enables probabilistic forecasting by mapping network outputs to probability distribution parameters instead of single point estimates.
- [Probabilistic Loss Functions](https://awesome-repositories.com/f/artificial-intelligence-ml/probabilistic-loss-functions.md) — Implements loss functions that map network outputs to probability distribution parameters for uncertainty quantification. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/tutorials/building.html))
- [Quantile Regression](https://awesome-repositories.com/f/artificial-intelligence-ml/regression-analysis/robust-regression/quantile-regression.md) — Produces non-parametric probabilistic forecasts by predicting specific target distribution percentiles using pinball loss.
- [Time Series Feature Engineering](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-feature-engineering.md) — Prepares raw tabular data for neural networks through automated scaling, normalization, and creation of training subsequences.
- [Time Series Forecasting](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting.md) — A comprehensive deep learning framework for building and training PyTorch-based time series forecasting architectures.
- [Dynamic Covariate Integration](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting/dynamic-covariate-integration.md) — Integrates static metadata and time-varying external variables into the model to improve prediction accuracy.
- [Multivariate Forecasting](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting/multivariate-forecasting.md) — Predicts multiple interdependent time series variables simultaneously while incorporating external covariates.
- [Probabilistic Forecasting](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting/probabilistic-forecasting.md) — Predicts future demand with uncertainty intervals by generating probability distributions instead of point estimates.
- [Distributed Training](https://awesome-repositories.com/f/artificial-intelligence-ml/distributed-training.md) — Supports spreading training processes across multiple CPUs or GPUs to accelerate model convergence. ([source](https://pytorch-forecasting.readthedocs.io))
- [Covariate Integration](https://awesome-repositories.com/f/artificial-intelligence-ml/linear-regression-models/covariance-aware-attribution/covariate-integration.md) — Incorporates static metadata and time-varying external variables into models to increase forecasting precision. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/tutorials/building.html))
- [Forecasting Interpretation Tools](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/model-evaluation-and-validation/model-evaluation-metrics/forecasting-interpretation-tools.md) — Analyzes forecasting performance using multi-horizon metrics, attention maps, and variable dependency charts to understand model behavior.
- [Hyperparameter Tuning](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/model-optimization-and-inference/training-algorithms/machine-learning-optimization/hyperparameter-tuning.md) — Optimizes model configuration settings using an automated search framework and pruning to improve training efficiency. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/CHANGELOG.html))
- [Multi-Horizon Evaluation Metrics](https://awesome-repositories.com/f/artificial-intelligence-ml/multi-horizon-evaluation-metrics.md) — Calculates performance metrics across various future time steps to measure accuracy over a prediction window.
- [Multi-Target Learning](https://awesome-repositories.com/f/artificial-intelligence-ml/multi-target-learning.md) — Simultaneously predicts multiple target variables using a combination of regression and classification. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/models.html))
- [Hyperparameter Optimizers](https://awesome-repositories.com/f/artificial-intelligence-ml/optimization-algorithms/hyperparameter-optimizers.md) — Offers an automated hyperparameter tuning and pruning framework to optimize deep learning architectures.
- [Prediction Visualization](https://awesome-repositories.com/f/artificial-intelligence-ml/prediction-visualization.md) — Generates actual-versus-prediction plots and dependency charts to interpret model behavior and performance. ([source](https://cdn.jsdelivr.net/gh/sktime/pytorch-forecasting@main/README.md))
- [Forecast Evaluation](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting/forecast-evaluation.md) — Provides multi-horizon time series metrics to measure forecasting accuracy across various prediction windows. ([source](https://cdn.jsdelivr.net/gh/sktime/pytorch-forecasting@main/README.md))
- [General Demand Forecasting](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting/intermittent-demand-forecasting/general-demand-forecasting.md) — Provides architectures specifically designed to handle complex time series patterns for demand prediction. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/tutorials.html))
- [ML-Based Forecasting Models](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting/ml-based-forecasting-models.md) — Provides specialized machine learning architectures for real-world forecasting with built-in interpretation capabilities. ([source](https://pytorch-forecasting.readthedocs.io))
- [Multivariate Quantile Prediction](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting/multivariate-forecasting/multivariate-quantile-prediction.md) — Produces probabilistic forecasts across multiple variables and long horizons using hierarchical sampling. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/tutorials.html))
- [Neural Basis Expansion Analysis](https://awesome-repositories.com/f/artificial-intelligence-ml/time-series-forecasting/neural-basis-expansion-analysis.md) — Provides interpretable time series forecasting through basis-function decomposition, specifically supporting architectures like N-BEATS.
- [Training Batch Generators](https://awesome-repositories.com/f/artificial-intelligence-ml/training-data-generation/training-batch-generators.md) — Samples overlapping encoder and decoder windows from time series to produce structured training batches. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/data.html))
- [Variable Dependency Analysis](https://awesome-repositories.com/f/artificial-intelligence-ml/variable-dependency-analysis.md) — Calculates partial dependency for specific variables to analyze their influence on forecasting predictions. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/CHANGELOG.html))

### Data & Databases

- [Time Series Tensor Pipelines](https://awesome-repositories.com/f/data-databases/time-series-tensor-pipelines.md) — Converts tabular time series data into PyTorch tensors while automating scaling and feature encoding.
- [Time Series Transformations](https://awesome-repositories.com/f/data-databases/time-series-transformations.md) — Transforms tabular time series data into normalized PyTorch tensors with configured encoder and decoder windows. ([source](https://pytorch-forecasting.readthedocs.io))
- [Time Series Data Loading](https://awesome-repositories.com/f/data-databases/time-series-data-loading.md) — Converts data tables into tensors while automating variable scaling, target normalization, and feature encoding. ([source](https://cdn.jsdelivr.net/gh/sktime/pytorch-forecasting@main/README.md))
- [Time Series Data Normalization](https://awesome-repositories.com/f/data-databases/time-series-data-normalization.md) — Scales input data using customizable encoder methods and robust scaling via specific quantiles. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/CHANGELOG.html))

### Part of an Awesome List

- [Interpretable](https://awesome-repositories.com/f/awesome-lists/ai/forecasting-models/interpretable.md) — Generates explainable forecasts using basis function decomposition to clarify model behavior. ([source](https://pytorch-forecasting.readthedocs.io/en/latest/tutorials.html))
- [Time Series Analysis](https://awesome-repositories.com/f/awesome-lists/data/time-series-analysis.md) — PyTorch-based library for neural network forecasting.

### Education & Learning Resources

- [Sliding Window Algorithms](https://awesome-repositories.com/f/education-learning-resources/sliding-window-algorithms.md) — Generates overlapping encoder and decoder windows from long time series to create training batches.