Code for IoTJ 2024 paper "SageFormer: Series-Aware Framework for Long-Term Multivariate Time-Series Forecasting".

Resources about time series forecasting and deep learning.

Time-Series Work Summary in CS Top Conferences (NIPS, ICML, ICLR, KDD, AAAI, WWW, IJCAI, CIKM, ICDM, ICDE, etc.)

UniT: Unified Geometry Learning with Group Autoregressive Transformer

[PVLDB 2024 Best Paper Nomination] TFB: Towards Comprehensive and Fair Benchmarking of Time Series Forecasting Methods

[KDD'24] Official code for our paper "Heterogeneity-Informed Meta-Parameter Learning for Spatiotemporal Time Series Forecasting".

AirLLM 70B inference with single 4GB GPU

A Library for Dynamic Graph Learning (NeurIPS 2023)

Official implementation of the paper "Frequency-domain MLPs are More Effective Learners in Time Series Forecasting"

[ICLR 2026] Official implementation for "Enhancing Multivariate Time Series Forecasting with Global Temporal Retrieval"

[ICLR 2024] Official implementation of "TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting"

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

collection of articles about PhD life written in 🇯🇵

This is the official code and supplementary materials for our AAAI-2024 paper: MASTER: Market-Guided Stock Transformer for Stock Price Forecasting. MASTER is a stock transformer for stock price for…

Official code, datasets and checkpoints for "Timer: Generative Pre-trained Transformers Are Large Time Series Models" (ICML 2024) and subsequent works

[AAAI2023] A PyTorch implementation of PDFormer: Propagation Delay-aware Dynamic Long-range Transformer for Traffic Flow Prediction.

[NeurIPS-25] How Different from the Past? Spatio-Temporal Time Series Forecasting with Self-Supervised Deviation Learning

[AAAI'26] Code for paper "Meta Dynamic Graph for Traffic Flow Prediction".

科研写作助手 (Research Writing Assistant)

FreqCycle boosts time series forecasting by jointly modeling low-frequency patterns and enhanced mid to high frequency features through novel time-frequency domain learning, achieving state-of-the…

[ICML 2024] Official implementation of: "Revitalizing Multivariate Time Series Forecasting: Learnable Decomposition with Inter-Series Dependencies and Intra-Series Variations Modeling".

This repo is the official implementation of "MART: MultiscAle Relational Transformer Networks for Trajectory Prediction", ECCV 2024.

Official implementation of our ICLR 2023 paper "Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting"

Pedestrian Crossing Intention Prediction with Feature Fusion and Spatio-Temporal Attention.