Skip to content

The entmax mapping and its loss, a family of sparse softmax alternatives.

active 2023-08-152026-03-07 (UTC)

Complete coverage26,356 / 26,356 hourly files (100%) · 2 absent upstream2023-08-152026-08-17 (UTC)
Events
185
Pushes
41
Pull requests
8
Issues
4
Stars
80
Forks
8

Activity over time

Daily event counts in the loaded window

Line chart, 936 days from 2023-08-15 to 2026-03-07. Pushes: 41 total, peak 11 in a day. Pull requests: 8 total, peak 2 in a day. Issues: 4 total, peak 2 in a day. Comments: 21 total, peak 5 in a day. Stars: 80 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
bpopeters6639615
andre-martins4220
12073822253002
vene2002
ssantos972001
R-N1001
weberBen1000
azhx1000

Recent activity

Latest issues, pull requests and releases

  • Issue comment#3912073822252024-12-21 09:50
    Use the automatically updated alpha endmax, where alpha is updated to a value less than 1 or a negative value
  • Issue comment#38bpopeters2024-12-19 16:31
    Sparsemax fails with torch.use_deterministic_algorithms(True) | deterministic CUDA cumsum
  • Issue comment#39bpopeters2024-12-19 15:53
    Use the automatically updated alpha endmax, where alpha is updated to a value less than 1 or a negative value
  • Issue comment#3612073822252024-12-18 18:08
    Instability of entmax bisect around 0?
  • Issue#3912073822252024-12-18 17:46
    Use the automatically updated alpha endmax, where alpha is updated to a value less than 1 or a negative value
  • Issue#38weberBen2024-12-18 17:08
    Sparsemax fails with torch.use_deterministic_algorithms(True) | deterministic CUDA cumsum
  • Issue comment#31bpopeters2024-05-23 14:39
    Add support for automatic mixed precision
  • Pull request#37bpopeters2024-05-21 14:31
  • Issue comment#37vene2024-05-21 14:10
    Enable returning support size with exact algorithms for alpha=1.5 and alpha=2
  • Issue comment#37bpopeters2024-05-20 14:30
    Enable returning support size with exact algorithms for alpha=1.5 and alpha=2
  • Issue comment#37vene2024-04-15 17:04
    Enable returning support size with exact algorithms for alpha=1.5 and alpha=2
  • Pull request#37bpopeters2024-04-15 16:28
  • Issue comment#35bpopeters2024-04-15 14:21
    Nans gradients for tensors with singleton dimensions for budget_bisect.
  • Issue#36azhx2024-04-10 20:46
    Instability of entmax bisect around 0?
  • Issue comment#35bpopeters2024-03-06 17:08
    Nans gradients for tensors with singleton dimensions for budget_bisect.
  • Issue comment#35ssantos972024-03-06 13:19
    Nans gradients for tensors with singleton dimensions for budget_bisect.
  • Issue comment#35bpopeters2024-03-06 12:22
    Nans gradients for tensors with singleton dimensions for budget_bisect.
  • Issue comment#35bpopeters2024-03-06 11:01
    Nans gradients for tensors with singleton dimensions for budget_bisect.
  • Issue#35ssantos972024-02-28 11:35
    Nans gradients for tensors with singleton dimensions.
  • Releasebpopeters2024-02-07 10:04
    entmax 1.3
  • Pull request#33bpopeters2024-02-07 09:51
  • Pull request#34bpopeters2024-02-06 13:46
  • Pull request#33andre-martins2024-02-04 20:02
  • Releasebpopeters2024-01-05 11:21
    entmax 1.2
  • Pull request#32bpopeters2024-01-04 15:12

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 80 stars here means stars gained during the window, not the repo's star count.