10 Pull requests merged by 2 people

• [Feature] LLM collector

  #2879 merged Apr 4, 2025

• [Feature] History API

  #2890 merged Apr 4, 2025

• [BugFix] Fix compile compatibility of PPO losses

  #2889 merged Apr 3, 2025

• [Feature] Pass lists of policy_factory

  #2888 merged Apr 3, 2025

• [Refactor] Fix repeats order

  #2887 merged Apr 3, 2025

• [Test] Fix warnings in tests

  #2886 merged Apr 3, 2025

• [BugFix] Fix .item() warning on tensors that require grad

  #2885 merged Apr 3, 2025

• [Feature] Support lazy tensordict inputs in ppo loss

  #2883 merged Apr 2, 2025

• [Refactor] MaskedCategorical cross_entropy usage for faster loss

  #2882 merged Apr 2, 2025

• [Refactor] Avoid padding in transformer wrapper

  #2881 merged Apr 2, 2025

2 Pull requests opened by 1 person

• [Feature] Support lazy tensordict inputs in KL reward transform

  #2884 opened Apr 2, 2025

• [Feature] More options for LLM collectors

  #2891 opened Apr 4, 2025

1 Unresolved conversation

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• No example for UnityMLAgentsEnv or Wrapper for single or multiagent training

  #2781 commented on Apr 4, 2025 • 0 new comments