#11 · Primary category： Time Series Machine Learning
neuralforecast
Scalable and user friendly neural :brain: forecasting algorithms.
Project last updated：08/24/26
GitHub Stars
4.3K
Forks
500
Contributors
69
License
Apache-2.0
Why we included this project
Forecasting teams that have outgrown ARIMA or ETS will find a practical upgrade in NeuralForecast. It's a PyTorch library with more than 30 neural models behind a single, consistent API, from older RNN and MLP designs to modern transformers like PatchTST, TFT, and iTransformer. Nixtla maintains official implementations of NBEATSx and NHITS, which helps when you want to compare against published results. The library outputs probabilistic forecasts, accepts exogenous and static covariates, and exposes interpretability hooks for trend and seasonality. The sklearn-style fit/predict interface and built-in tuning with Ray or Optuna let a small team run model comparisons without building a training pipeline from scratch.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
netdata
The fastest path to AI-powered full stack observability, even for lean teams.
catboost
A fast, scalable, high performance Gradient Boosting on Decision Trees library, used for ranking, classification, regression and other machine learning tasks for Python, R, Java, C++. Supports computation on CPU and GPU.
sktime
A unified framework for machine learning with time series
darts
A python library for user-friendly forecasting and anomaly detection on time series.
featuretools
An open source python library for automated feature engineering