[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-9810":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":24,"hasPages":24,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":46,"readmeContent":47,"aiSummary":48,"trendingCount":16,"starSnapshotCount":16,"syncStatus":17,"lastSyncTime":49,"discoverSource":50},9810,"neural_prophet","ourownstory\u002Fneural_prophet","ourownstory","NeuralProphet: A simple forecasting package","https:\u002F\u002Fneuralprophet.com",null,"Python",4284,513,55,73,0,2,13,1,30.13,"MIT License",false,"main",true,[26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45],"artificial-intelligence","autoregression","deep-learning","fbprophet","forecast","forecasting","forecasting-algorithm","forecasting-model","machine-learning","neural","neural-network","neuralprophet","prediction","prophet","python","pytorch","seasonality","time-series","timeseries","trend","2026-06-12 02:02:12","[![GitHub release (latest SemVer)](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fv\u002Frelease\u002Fourownstory\u002Fneural_prophet?logo=github)](https:\u002F\u002Fgithub.com\u002Fourownstory\u002Fneural_prophet\u002Freleases)\n[![Pypi_Version](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fneuralprophet.svg)](https:\u002F\u002Fpypi.python.org\u002Fpypi\u002Fneuralprophet)\n[![Python Version](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fpython-3.9+-blue?logo=python)](https:\u002F\u002Fwww.python.org\u002F)\n[![Code style: black](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fcode%20style-black-000000.svg)](https:\u002F\u002Fgithub.com\u002Fpsf\u002Fblack)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-MIT-brightgreen)](https:\u002F\u002Fopensource.org\u002Flicenses\u002FMIT)\n[![Tests](https:\u002F\u002Fgithub.com\u002Fourownstory\u002Fneural_prophet\u002Factions\u002Fworkflows\u002Ftests.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002Fourownstory\u002Fneural_prophet\u002Factions\u002Fworkflows\u002Ftests.yml)\n[![codecov](https:\u002F\u002Fcodecov.io\u002Fgh\u002Fourownstory\u002Fneural_prophet\u002Fbranch\u002Fmaster\u002Fgraph\u002Fbadge.svg?token=U5KXCL55DW)](https:\u002F\u002Fcodecov.io\u002Fgh\u002Fourownstory\u002Fneural_prophet)\n[![Slack](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fslack-@neuralprophet-CF0E5B.svg?logo=slack&logoColor=white&labelColor=3F0E40)](https:\u002F\u002Fneuralprophet.slack.com\u002Fjoin\u002Fshared_invite\u002Fzt-sgme2rw3-3dCH3YJ_wgg01IXHoYaeCg#\u002Fshared-invite\u002Femail)\n[![Downloads](https:\u002F\u002Fstatic.pepy.tech\u002Fpersonalized-badge\u002Fneuralprophet?period=total&units=international_system&left_color=black&right_color=blue&left_text=Downloads)](https:\u002F\u002Fpepy.tech\u002Fproject\u002Fneuralprophet)\n\n![NP-logo-wide_cut](https:\u002F\u002Fuser-images.githubusercontent.com\u002F21246060\u002F111388960-6c367e80-866d-11eb-91c1-46f2c0d21879.PNG)\n\n\nPlease note that the project is still in beta phase. Please report any issues you encounter or suggestions you have. We will do our best to address them quickly. Contributions are very welcome!\n\n# NeuralProphet: human-centered forecasting\nNeuralProphet is an easy to learn framework for interpretable time series forecasting.\nNeuralProphet is built on PyTorch and combines Neural Networks and traditional time-series algorithms, inspired by [Facebook Prophet](https:\u002F\u002Fgithub.com\u002Ffacebook\u002Fprophet) and [AR-Net](https:\u002F\u002Fgithub.com\u002Fourownstory\u002FAR-Net).\n- With a few lines of code, you can define, customize, visualize, and evaluate your own forecasting models.\n- It is designed for iterative human-in-the-loop model building. That means that you can build a first model quickly, interpret the results, improve, repeat. Due to the focus on interpretability and customization-ability, NeuralProphet may not be the most accurate model out-of-the-box; so, don't hesitate to adjust and iterate until you like your results.\n- NeuralProphet is best suited for time series data that is of higher-frequency (sub-daily) and longer duration (at least two full periods\u002Fyears).\n\n\n## Documentation\nThe [documentation page](https:\u002F\u002Fneuralprophet.com) may not be entirely up to date. Docstrings should be reliable, please refer to those when in doubt. We are working on an improved documentation. We appreciate any help to improve and update the docs.\n\nFor a visual introduction to NeuralProphet, [view this presentation](notes\u002FNeuralProphet_Introduction.pdf).\n\n## Contribute\nWe compiled a [Contributing to NeuralProphet](CONTRIBUTING.md) page with practical instructions and further resources to help you become part of the family. \n\n## Community\n#### Discussion and Help\nIf you have any questions or suggestion, you can participate in [our community right here on Github](https:\u002F\u002Fgithub.com\u002Fourownstory\u002Fneural_prophet\u002Fdiscussions)\n\n#### Slack Chat\nWe also have an active [Slack community](https:\u002F\u002Fjoin.slack.com\u002Ft\u002Fneuralprophet\u002Fshared_invite\u002Fzt-sgme2rw3-3dCH3YJ_wgg01IXHoYaeCg). Come and join the conversation!\n\n## Tutorials\n[![Open All Collab](https:\u002F\u002Fcolab.research.google.com\u002Fassets\u002Fcolab-badge.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fourownstory\u002Fneural_prophet)\n\nThere are several [example notebooks](docs\u002Fsource\u002Ftutorials) to help you get started. \n\nYou can find the datasets used in the tutorials, including data preprocessing examples, in our [neuralprophet-data repository](https:\u002F\u002Fgithub.com\u002Fourownstory\u002Fneuralprophet-data).\n\nPlease refer to our [documentation page](https:\u002F\u002Fneuralprophet.com) for more resources.\n\n### Minimal example\n```python\nfrom neuralprophet import NeuralProphet\n```\nAfter importing the package, you can use NeuralProphet in your code:\n```python\nm = NeuralProphet()\nmetrics = m.fit(df)\nforecast = m.predict(df)\n```\nYou can visualize your results with the inbuilt plotting functions:\n```python\nfig_forecast = m.plot(forecast)\nfig_components = m.plot_components(forecast)\nfig_model = m.plot_parameters()\n```\nIf you want to forecast into the unknown future, extend the dataframe before predicting:\n```python\nm = NeuralProphet().fit(df, freq=\"D\")\ndf_future = m.make_future_dataframe(df, periods=30)\nforecast = m.predict(df_future)\nfig_forecast = m.plot(forecast)\n```\n## Install\nYou can now install neuralprophet directly with pip:\n```shell\npip install neuralprophet\n```\n\n### Install options\n\nIf you plan to use the package in a Jupyter notebook, we recommended to install the 'live' version:\n```shell\npip install neuralprophet[live]\n```\nThis will allow you to enable `plot_live_loss` in the `fit` function to get a live plot of train (and validation) loss.\n\nIf you would like the most up to date version, you can instead install directly from github:\n```shell\ngit clone \u003Ccopied link from github>\ncd neural_prophet\npip install .\n```\n\nNote for Windows users: Please use WSL2.\n\n## Features\n### Model components\n* Autoregression: Autocorrelation modelling - linear or NN (AR-Net).\n* Trend: Piecewise linear trend with optional automatic changepoint detection.\n* Seasonality: Fourier terms at different periods such as yearly, daily, weekly, hourly.\n* Lagged regressors: Lagged observations (e.g temperature sensor) - linear or NN.\n* Future regressors: In advance known features (e.g. temperature forecast) - linear or NN.\n* Events: Country holidays & recurring custom events.\n* Global Modeling: Components can be local, global or 'glocal' (global + regularized local)\n\n\n### Framework features\n* Multiple time series: Fit a global\u002Fglocal model with (partially) shared model parameters.\n* Uncertainty: Estimate values of specific quantiles - Quantile Regression.\n* Regularize modelling components.\n* Plotting of forecast components, model coefficients and more.\n* Time series crossvalidation utility.\n* Model checkpointing and validation.\n\n\n### Coming soon\u003Csup>:tm:\u003C\u002Fsup>\n\n* Cross-relation of lagged regressors.\n* Static metadata regression for multiple series\n* Logistic growth for trend component.\n\nFor a list of past changes, please refer to the [releases page](https:\u002F\u002Fgithub.com\u002Fourownstory\u002Fneural_prophet\u002Freleases).\n\n## Cite\nPlease cite [NeuralProphet](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.15397) in your publications if it helps your research:\n```\n@misc{triebe2021neuralprophet,\n      title={NeuralProphet: Explainable Forecasting at Scale}, \n      author={Oskar Triebe and Hansika Hewamalage and Polina Pilyugina and Nikolay Laptev and Christoph Bergmeir and Ram Rajagopal},\n      year={2021},\n      eprint={2111.15397},\n      archivePrefix={arXiv},\n      primaryClass={cs.LG}\n}\n```\n### Many Thanks To Our Contributors:\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fourownstory\u002Fneural_prophet\u002Fgraphs\u002Fcontributors\">\n  \u003Cimg src=\"https:\u002F\u002Fcontrib.rocks\u002Fimage?repo=ourownstory\u002Fneural_prophet\" \u002F>\n\u003C\u002Fa>\n\n## About\nNeuralProphet is an open-source community project, supported by awesome people like you. \nIf you are interested in joining the project, please feel free to reach out to me (Oskar) - you can find my email on the [NeuralProphet Paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.15397).\n","NeuralProphet 是一个简单易用的时间序列预测框架。它基于 PyTorch 构建，结合了神经网络与传统时间序列算法，受到 Facebook Prophet 和 AR-Net 的启发。其核心功能包括通过几行代码即可定义、定制、可视化和评估预测模型，并且特别强调可解释性和迭代优化过程。NeuralProphet 适用于高频次（如每日以下）且持续时间较长（至少两个完整周期\u002F年）的时间序列数据场景，非常适合需要快速构建初步模型并进行逐步改进的用户。","2026-06-11 03:24:51","top_topic"]