#135135th of 214 in ML frameworks and librariesFor data and documents, data analysis
neuralforecast
Python library of 30+ neural time-series forecasting models with sklearn-style API
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.3k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Oct 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
NeuralForecast is a Python library that implements neural forecasting models, from MLP, LSTM and TCN to NHITS, NBEATSx, TFT, PatchTST and iTransformer, behind a fit/predict interface. It supports exogenous variables, static covariates, probabilistic forecasts and hyperparameter tuning through Ray or Optuna. It shares input and output formats with StatsForecast, MLForecast and HierarchicalForecast.
Who it is for: Data scientists and ML engineers building Python forecasting pipelines
Strengths
- Over 30 forecasting models behind one fit/predict API
- Official NHITS and NBEATSx implementations from the papers' authors
- Auto models tune hyperparameters with Ray or Optuna
- Since 3.3.0, saves safetensors plus JSON, so loading runs no code
Weaknesses
- Models saved before 3.3.0 need allow_pickle=True or a migration step
- TimeLLM cannot be saved or loaded; train and predict in one process
- Custom losses, optimizers and schedulers must be registered before save and load
- Time-series forecasting only; no general LLM or chat features
What it needs
- Needs Ray, Optuna, utilsforecast, coreforecast
- Models: NBEATS, NBEATSx, NHITS, TFT, PatchTST
Also in ML frameworks and libraries
See all 214| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | ultralyticsPython library and CLI for training and running YOLO vision models | 94 out of 100 |
| 3 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 94 out of 100 |
| #4 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |