Skip to content

时间序列预测 [ICLR 2024] Official implementation of "TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting"

active 2024-03-052026-07-04 (UTC)

Partial coverage20,372 / 26,126 hourly files (78%) · 2 absent upstream · 5,751 failed, retryable2023-08-202026-08-13 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
1.5K
Pushes
30
Pull requests
2
Issues
134
Stars
1.1K
Forks
127

Activity over time

Daily event counts in the loaded window

Line chart, 852 days from 2024-03-05 to 2026-07-04. Pushes: 30 total, peak 3 in a day. Pull requests: 2 total, peak 1 in a day. Issues: 134 total, peak 4 in a day. Comments: 154 total, peak 8 in a day. Stars: 1,075 total, peak 29 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

ContributorContributionsPushesPRsComments
kwuking14430070
kevinliu2000210015
DirectorZ17004
liangzrtvjivo5004
13539182854001
lhhhappy4000
rbm1234564003
yslbit4003
carusyte3011
qq234567890lyj3001
lou-yimin3003
Caominjun93002
dongdong20233010
Diroooomy3001
warmful-sun3001
lzmax8883001
zmce20183000
hczs3001
guanfj3002
Xin-Ray3002

Recent activity

Latest issues, pull requests and releases

  • Issue#197T2zxy2026-03-14 18:37
    我是一个新手,对于论文中的图片复现不是很懂如何实现
  • Issue#196T2zxy2026-02-02 02:04
    复现疑惑
  • Issue#136kevinliu20002026-02-01 14:55
    Add Type Annotations to Improve Code Quality and Documentation
  • Issue comment#146kevinliu20002026-01-04 11:52
    FEATURE REQUEST: Time Series Data Augmentation Module
  • Issue comment#154kevinliu20002026-01-04 11:49
    Enhancing Model Interpretability through Feature Attribution Mechanisms
  • Issue comment#194kevinliu20002026-01-04 11:48
    Questions Regarding Classification Task Reproduction for TimeMixer++
  • Issue#1941st-hoan2025-12-19 09:34
    Questions Regarding Classification Task Reproduction for TimeMixer++
  • Issue comment#158kevinliu20002025-12-11 03:41
    Why is input normalization (mean/std) needed after scaling? Performance drops if removed
  • Issue#192kevinliu20002025-12-10 06:44
    超参数设置问题
  • Issue#193erikalien55952025-12-07 01:21
    PEMS数据集预测问题
  • Issue comment#191kevinliu20002025-12-05 07:51
    维度不一致:Dimensions do not match
  • Issue comment#155kevinliu20002025-12-05 07:47
    bug: missing complexity analysis
  • Issue comment#142kevinliu20002025-12-05 07:45
    Enhance Checkpoint Management and Model Serialization
  • Issue comment#143kevinliu20002025-12-05 07:43
    GPU Memory Optimization Needed
  • Issue comment#160kevinliu20002025-12-05 07:39
    Support for Weights & Biases Integration
  • Issue comment#190kevinliu20002025-12-05 07:34
    x_mark 使用repeat 还是 repeat_interleave?
  • Issue comment#162kevinliu20002025-10-17 06:35
    Implement Flexible Data Caching Mechanism
  • Issue comment#168kevinliu20002025-10-17 06:33
    Improve Future Temporal Features Implementation
  • Issue comment#144kevinliu20002025-10-17 06:29
    We need docker support! 🐳
  • Issue#169kevinliu20002025-10-17 03:55
    Implement Model Explainability Framework for TimeMixer
  • Issue comment#169kevinliu20002025-10-17 03:54
    Implement Model Explainability Framework for TimeMixer
  • Issue#134kevinliu20002025-10-16 14:48
    Timemixer++论文中的一些困惑点
  • Issue comment#189kevinliu20002025-10-16 14:48
    How to use TimeMixer++
  • Issue#181kevinliu20002025-10-16 10:37
    Meaning of label_len
  • Issue#182kevinliu20002025-10-16 10:36
    x_dec is not even used in Model.forecast

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 1,075 stars here means stars gained during the window, not the repo's star count.