Commits

Breadcrumbs

History for

onmain

User selector

All users

Datepicker

All time

Commit History

Commits on Dec 2, 2024

• [BugFix] Better account of composite distributions in PPO

  Show description for 90c8e40
  vmoens

  committed
  90c8e40

Commits on Nov 18, 2024

• [Doc] Fix broken links and formatting issues in doc

  Show description for 5a2d9e2
  vmoens

  committed
  5a2d9e2

Commits on Nov 6, 2024

• [BugFix] Support for tensor collection in the PPOLoss (#2543)

  Show description for 0eabb78
  

  priba

  and

  Pau Riba

  authored
  0eabb78

Commits on Sep 17, 2024

• [Feature] Make benchmarked losses compatible with torch.compile

  Show description for 2332909
  vmoens

  committed
  2332909

Commits on Sep 12, 2024

• [BugFix] compatibility to new Composite dist log_prob/entropy APIs

  Show description for 36545af
  vmoens

  committed
  36545af

Commits on Sep 4, 2024

• [BugFix] Allow for composite action distributions in PPO/A2C losses (#2391)

  albertbou92

  authored
  49d7f74

Commits on Aug 13, 2024

• [Deprecation] Deprecate default num_cells in MLP (#2395)

  vmoens

  authored
  25e8bd2

Commits on Aug 7, 2024

• [Refactor] Rename specs to simpler names (#2368)

  vmoens

  committed
  607db8b

Commits on Aug 4, 2024

• [Refactor] Refactor calls to get without default that raise KeyError (#2353)

  vmoens

  authored
  35a1c5b

Commits on Aug 2, 2024

• [Docs] Use more appropriate ActorValueOperator in PPOLoss documentation (#2350)

  GaetanLepage

  authored
  59d2ae1

Commits on Jul 25, 2024

• [Refactor] Deprecate NormalParamWrapper (#2308)

  vmoens

  authored
  474e837

Commits on Jul 23, 2024

• [Feature] Construct loss by providing lists of modules

  Show description for f840a1a
  vmoens

  committed
  f840a1a

Commits on May 29, 2024

• [Feature] Add KL approximation in PPO loss metadata (#2166)

  albertbou92

  authored
  767a877

Commits on May 2, 2024

• [Feature] Autocomplete for losses (#2148)

  vmoens

  authored
  6822468

Commits on Apr 25, 2024

• [Versioning] Deprecations for 0.4 (#2109)

  vmoens

  authored
  36e1309

Commits on Apr 2, 2024

• [BugFix] Set default value for normalize_advantage to False. (#2050)

  DobromirM

  authored
  2461eb2

Commits on Mar 19, 2024

• [BugFix] Fix clip_fraction in PO losses (#2021)

  vmoens

  authored
  4bce371

Commits on Mar 18, 2024

• [Feature] Adds value clipping in ClipPPOLoss loss (#2005)

  Show description for 43c6bca
  

  albertbou92

  and

  vmoens

  authored
  43c6bca

Commits on Feb 27, 2024

• [BugFix] Dedicated tests for on policy losses reduction parameter (#1974)

  Show description for 3d65083
  

  albertbou92

  and

  vmoens

  authored
  3d65083

• [BugFix] Refactor reductions (#1968)

  vmoens

  authored
  db4ad23

Commits on Feb 17, 2024

• [BugFix] Fix KLPENPPOLoss KL computation (#1922)

  vmoens

  authored
  e538fdc

Commits on Feb 15, 2024

• [Feature] Add reduction parameter to On-Policy losses. (#1890)

  Show description for 67f659c
  

  albertbou92

  and

  vmoens

  authored
  67f659c

Commits on Jan 31, 2024

• [Deprecation] Deprecate in prep for release (#1820)

  vmoens

  authored
  b5e90c4

Commits on Jan 30, 2024

• [Feature] Fine control over devices in collectors (#1835)

  vmoens

  authored
  b1cc796

Commits on Jan 29, 2024

• [Doc] Make tutos runnable without colab (#1826)

  vmoens

  authored
  6277226

Commits on Jan 23, 2024

• [Doc, Feature] Doc improvements for video recording and CSV video formats (#1829)

  vmoens

  authored
  24d14ad

• [Feature] Non-functional objectives (PPO, A2C, Reinforce) (#1804)

  vmoens

  authored
  5b67dd3

Commits on Jan 19, 2024

• [Feature] TorchRL2Gym conversion (#1795)

  vmoens

  authored
  c3ffb5a

Commits on Nov 24, 2023

• [Refactor] Refactor functional calls in losses (#1707)

  vmoens

  authored
  bc7595f

Commits on Nov 23, 2023

• [Algorithm] IMPALA and VTrace module (#1506)

  Show description for b38d4b7
  

  albertbou92

  and

  vmoens

  authored
  b38d4b7

Commits on Oct 5, 2023

• [Algorithm] RLHF end-to-end, clean (#1597)

  Show description for fe19cf5
  

  3 peopleauthored
  fe19cf5

Commits on Oct 2, 2023

• [Feature] Make advantages compatible with Terminated, Truncated, Done (#1581)

  Show description for 106368f
  

  3 peopleauthored
  106368f

Commits on Jul 26, 2023

• [Refactor] Remove loss device (#1395)

  Show description for c06ed70
  matteobettini

  authored
  c06ed70

Commits on Jul 11, 2023

• [BugFix] Fix tutorials (#1382)

  vmoens

  authored
  5a3f9e0

Commits on Jul 2, 2023

• [BugFix] Fix brax (#1346)

  vmoens

  authored
  fa4fe1d

Pagination

PreviousNext