Projects sharing features with SCINet

Prophet is a time series forecasting library and decomposition tool that uses an additive regression model to predict future values. It functions as an uncertainty estimation tool, calculating confidence intervals and error metrics to quantify the risk associated with future predictions. The project is distinguished by its ability to incorporate human-interpretable parameters for model tuning and its use of Bayesian inference for parameter estimation. It supports the integration of external regressors and special event modeling to account for the impact of holidays and specific dates on forec

Prophet is a predictive analytics framework and time series regression library designed for forecasting future values. It uses additive models to fit non-linear growth and periodic seasonal patterns, providing tools for producing forecasts with integrated error measurement. The project handles multiple seasonalities and holiday effects to improve accuracy for periodic data. It supports the integration of external regressors and manages data irregularities, such as missing data and outliers, to maintain prediction stability. The framework covers a broad range of analysis capabilities, includi

FiLM: Frequency improved Legendre Memory Model for Long-term Time Series Forecasting https://arxiv.org/abs/2205.08897

This repo is the implementation of our manuscript entitled AirFormer: Predicting Nationwide Air Quality in China with Transformers. The code is based on Pytorch 1.10, and tested on Ubuntu 16.04 with a NVIDIA RTX A6000 GPU with 48 GB memory.