Skip to content

TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecasting

active 2024-03-192026-01-25 (UTC)

Partial coverage17,631 / 22,640 hourly files (78%) · 2 absent upstream · 5,008 failed, retryable2024-01-122026-08-12 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
150
Pushes
1
Pull requests
0
Issues
6
Stars
111
Forks
13

Activity over time

Daily event counts in the loaded window

Line chart, 678 days from 2024-03-19 to 2026-01-25. Pushes: 1 total, peak 1 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 6 total, peak 1 in a day. Comments: 19 total, peak 3 in a day. Stars: 111 total, peak 4 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
Atik-Ahamed10109
liuquangao4003
hbwzj2002
xiyang20022002
largraf1001
zjy-cqu1000
wuzzeze1000
Jerry-a-rookie1000
faceWelcome1001
Dreamhepeng1000
Jimmy-76641001
shandongpengyuyan1000

Recent activity

Latest issues, pull requests and releases

  • Issue comment#15Atik-Ahamed2025-08-22 12:14
    About the Memory footprint (in GB) figures
  • Issue#15Jerry-a-rookie2025-07-09 03:18
    About the Memory footprint (in GB) figures
  • Issue#13zjy-cqu2024-11-11 01:27
    mse or mae is either too big or too small
  • Issue comment#8largraf2024-09-29 16:58
    not able to achieve the same excellent results
  • Issue comment#8Atik-Ahamed2024-09-12 21:16
    not able to achieve the same excellent results
  • Issue comment#12Atik-Ahamed2024-09-07 20:33
    where to download the data?
  • Issue comment#11Atik-Ahamed2024-09-07 20:31
    Can TimeMachine be applied to multivariate time series classification(MTSC)?
  • Issue#11wuzzeze2024-09-04 01:33
    Can TimeMachine be applied to multivariate time series classification(MTSC)?
  • Issue comment#10Atik-Ahamed2024-09-02 17:24
    Do you only apply transpose MAMBA under channel independent settings?
  • Issue#10liuquangao2024-08-17 05:29
    Do you only apply transpose MAMBA under channel independent settings?
  • Issue comment#8liuquangao2024-08-15 12:20
    not able to achieve the same excellent results
  • Issue comment#8liuquangao2024-08-15 12:18
    not able to achieve the same excellent results
  • Issue comment#8liuquangao2024-08-15 12:17
    not able to achieve the same excellent results
  • Issue comment#9Atik-Ahamed2024-07-10 22:41
    What data shape could be mamba's correct input?
  • Issue comment#7hbwzj2024-06-01 17:37
    问题求助
  • Issue comment#8xiyang20022024-05-30 04:44
    not able to achieve the same excellent results
  • Issue comment#8xiyang20022024-05-28 05:22
    not able to achieve the same excellent results
  • Issue comment#8Atik-Ahamed2024-05-27 18:57
    not able to achieve the same excellent results
  • Issue comment#7Atik-Ahamed2024-05-27 18:37
    问题求助
  • Issue comment#7hbwzj2024-05-22 05:06
    问题求助
  • Issue comment#6Atik-Ahamed2024-05-08 20:26
    Multi-GPU Execution Error: Invalid device id
  • Issue comment#4faceWelcome2024-04-11 03:39
    Nan loss!
  • Issue#3Dreamhepeng2024-04-08 04:07
    import selective_scan_cuda ImportError: /home/hp/miniforge3/envs/mamba/lib/python3.10/site-packages/selective_scan_cuda.cpython-310-x86_64-linux-gnu.so: undefined symbol: _ZN3c104cuda9SetDeviceEi
  • Issue comment#2Jimmy-76642024-04-03 03:41
    The training time of the model
  • Issue#1shandongpengyuyan2024-04-01 06:42
    How to install causal conv1d and mamba-sms

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 111 stars here means stars gained during the window, not the repo's star count.