#
Starred repositories
[KDD 2025] DUET: Dual Clustering Enhanced Multivariate Time Series Forecasting
[PVLDB 2024 Best Paper Nomination] TFB: Towards Comprehensive and Fair Benchmarking of Time Series Forecasting Methods
Official implementation of our ICLR 2023 paper "Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting"
PyTorch Implementation of "TimeFilter: Patch-Specific Spatial-Temporal Graph Filtration for Time Series Forecasting" (ICML 2025)
PyTorch Implementation of "Adaptive Multi-Scale Decomposition Framework for Time Series Forecasting" (AAAI2025)
This project demonstrates the computation process of the RWKV (Receptance Weighted Key Value) model through Excel spreadsheets.
14
Updated Apr 30, 2025
[ICLR 2025 Spotlight] Official implementation of "Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts"
ProbTS is a benchmarking toolkit for time series forecasting.
DeerFlow is a community-driven Deep Research framework, combining language models with tools like web search, crawling, and Python execution, while contributing back to the open-source community.
Open Source DeepWiki: AI-Powered Wiki Generator for GitHub/Gitlab/Bitbucket Repositories. Join the discord: https://discord.gg/gMwThUMeme
Papers and datasets for tensor time series.
Code for IoTJ 2024 paper "SageFormer: Series-Aware Framework for Long-Term Multivariate Time-Series Forecasting".
Video+code lecture on building nanoGPT from scratch
A nanoGPT pipeline packed in a spreadsheet
[TKDE 2025] RAGIC: Risk-Aware Generative Framework for Stock Interval Construction
2
Updated Feb 25, 2025
[ICAIF 2024] Adaptive and Explainable Margin Trading via Large Language Models on Portfolio Management
A complete ML pipeline that predicts next-day stock prices and tests simple trading strategies — powered by AlphaForge.
Repository for "FreDF: Learning to Forecast in the Transformed Domain"
Official implementation of the paper "FilterNet: Harnessing Frequency Filters for Time Series Forecasting"
An Informer-LSTM model for State-of-Charge Estimation of Lithium-Ion Batteries
PyTorch implementation of "Rethinking the Power of Timestamps for Robust Time Series Forecasting: A Global-Local Fusion Perspective" (NeurIPS 2024)