Code

Release 1.4.0#260

Code

Merged

Toni-SM merged 33 commits intomainToni-SM/skrl:mainfrom developToni-SM/skrl:developCopy head branch name to clipboardJan 16, 2025

+16,149-7,562

Commits

Commits on Sep 26, 2024

• Mixed double precision for PPO algorithm (#155)

  Show description for bacb0f5
  lopatovsky

  authored
  bacb0f5

  Copy full SHA for bacb0f5

Commits on Oct 12, 2024

• Call agent's pre-interaction during evaluation (#210)

  Show description for f585796
  Toni-SM

  authored
  f585796

  Copy full SHA for f585796

Commits on Oct 18, 2024

• Space-tensor conversion (#206)

  Show description for 934fbaa
  Toni-SM

  authored
  934fbaa

  Copy full SHA for 934fbaa

Commits on Nov 2, 2024

• Remove OpenAI Gym (gym) from dependencies and source code (#220)

  Show description for 7dc161e
  Toni-SM

  authored
  7dc161e

  Copy full SHA for 7dc161e

• Merge branch 'main' into develop

  Toni-SM

  committed
  ae4e09e

  Copy full SHA for ae4e09e

Commits on Nov 3, 2024

• Fix: with SAC, a new training batch should be sampled for each gradient step (#208)

  Show description for 5fce807
  

  3 peopleauthored
  5fce807

  Copy full SHA for 5fce807

• Fix Sampling inside gradient loop issue (#183)

  Show description for 9252ec9
  

  3 peopleauthored
  9252ec9

  Copy full SHA for 9252ec9

• Add class mapping of categorical model (#216)

  Show description for eff7295
  

  Telios

  and

  Toni-SM

  authored
  eff7295

  Copy full SHA for eff7295

Commits on Nov 4, 2024

• Update pre-commit hooks (#221)

  Show description for 88ac11f
  Toni-SM

  authored
  88ac11f

  Copy full SHA for 88ac11f

Commits on Nov 5, 2024

• Apply black and codespell pre-commit hooks (#222)

  Show description for 4db2956
  Toni-SM

  authored
  4db2956

  Copy full SHA for 4db2956

Commits on Nov 23, 2024

• Docs update (#228)

  Show description for bbe532d
  Toni-SM

  authored
  bbe532d

  Copy full SHA for bbe532d

Commits on Dec 2, 2024

• Improve model instantiators (#232)

  Show description for 6324e46
  Toni-SM

  authored
  6324e46

  Copy full SHA for 6324e46

• Disable torch distribution argument validation to improve performance (#233)

  Show description for 57f60df
  Toni-SM

  authored
  57f60df

  Copy full SHA for 57f60df

Commits on Dec 6, 2024

• Use ML framework specific device parsing in source code (#234)

  Show description for 87250fa
  Toni-SM

  authored
  87250fa

  Copy full SHA for 87250fa

Commits on Dec 8, 2024

• Replace PyTorch's BatchSampler by Python slice when sampling data from memory (#235)

  Show description for 3d4eb76
  Toni-SM

  authored
  3d4eb76

  Copy full SHA for 3d4eb76

Commits on Dec 19, 2024

• Update JAX installation warning note (#241)

  Toni-SM

  authored
  d2aee9f

  Copy full SHA for d2aee9f

Commits on Dec 21, 2024

• Shared model instantiator's default parameters (#242)

  Show description for 23b61dc
  Toni-SM

  authored
  23b61dc

  Copy full SHA for 23b61dc

Commits on Jan 1, 2025

• Automatic mixed precision training in PyTorch (#243)

  Show description for 2882db4
  Toni-SM

  authored
  2882db4

  Copy full SHA for 2882db4

• Fix SAC experiment directory key name (#244)

  Toni-SM

  authored
  f39aadc

  Copy full SHA for f39aadc

Commits on Jan 5, 2025

• Fix Optax's learning rate schedulers integration in JAX (#245)

  Toni-SM

  authored
  e49f98f

  Copy full SHA for e49f98f

• Isaac Lab wrapper's multi-agent state retrieval with gymnasium 1.0

  Toni-SM

  authored
  a7f82b2

  Copy full SHA for a7f82b2

• Add method to initialize lazy modules' parameters

  Toni-SM

  committed
  65da82b

  Copy full SHA for 65da82b

Commits on Jan 7, 2025

• Update examples that use model instantiators to the latest API and Initialize models' lazy modules (#247)

  Toni-SM

  authored
  deb28ff

  Copy full SHA for deb28ff

• Update environment loader and wrapper for Isaac Lab 2.0 (#248)

  Toni-SM

  authored
  95663f2

  Copy full SHA for 95663f2

Commits on Jan 8, 2025

• Update Gymnasium checking for vectorized environments (#250)

  Toni-SM

  authored
  e5c6b81

  Copy full SHA for e5c6b81

Commits on Jan 15, 2025

• Update AMP agent to use the environment's terminated and truncated data, and the KL Adaptive learning rate scheduler (#253)

  Toni-SM

  authored
  95b0e02

  Copy full SHA for 95b0e02

• Update runner implementations to support definition of arbitrary agents and their models (#254)

  Toni-SM

  authored
  16224b0

  Copy full SHA for 16224b0

• Fix multi-agent learning rate scheduler in JAX (#255)

  Toni-SM

  authored
  9bda6c5

  Copy full SHA for 9bda6c5

• Add automatic mixed precision support for multi-agent and deal torch.amp deprecation warning (#256)

  Show description for d11a020
  Toni-SM

  authored
  d11a020

  Copy full SHA for d11a020

Commits on Jan 16, 2025

• Update gymnasium make vector API (#257)

  Toni-SM

  authored
  cb21eba

  Copy full SHA for cb21eba

• Fix memory sampling when sequence_length is specified

  Toni-SM

  committed
  7f9992b

  Copy full SHA for 7f9992b

• Treat truncation signal when computing 'done' (environment reset) (#259)

  Toni-SM

  authored
  663f546

  Copy full SHA for 663f546

• Allow the use of the deterministic/stochastic actions during evaluation (#252)

  Toni-SM

  authored
  fdfb8a5

  Copy full SHA for fdfb8a5