Trinity-RFT

Tutorial

  • Quick Start
  • Off-Policy RFT
  • Asynchronous RFT
  • Multi-Turn RFT
  • Offline DPO
  • Data Processing
  • Configuration Guide
  • Developer Guide

API Reference

  • trinity.buffer package
  • trinity.explorer package
  • trinity.trainer package
    • Subpackages
      • trinity.trainer.verl package
    • Submodules
    • Module contents
  • trinity.manager package
  • trinity.common package
  • trinity.utils package
Trinity-RFT
  • trinity.trainer package
  • trinity.trainer.verl package
  • Edit on GitHub

trinity.trainer.verl package

Submodules

  • trinity.trainer.verl.core_algos module
  • trinity.trainer.verl.dp_actor module
    • DataParallelPPOActor
      • DataParallelPPOActor.__init__()
      • DataParallelPPOActor.set_algorithm()
      • DataParallelPPOActor.update_policy()
  • trinity.trainer.verl.fsdp_workers module
    • ActorRolloutRefWorker
      • ActorRolloutRefWorker.__init__()
      • ActorRolloutRefWorker.init_model()
      • ActorRolloutRefWorker.setup_weight_sync_group()
      • ActorRolloutRefWorker.sync_weight()
      • ActorRolloutRefWorker.upload_state_dict()
      • ActorRolloutRefWorker.set_algorithm()
      • ActorRolloutRefWorker.update_actor()
      • ActorRolloutRefWorker.compute_log_prob()
      • ActorRolloutRefWorker.compute_ref_log_prob()
      • ActorRolloutRefWorker.save_checkpoint()
      • ActorRolloutRefWorker.load_checkpoint()
      • ActorRolloutRefWorker.clear_optimizer_state()
      • ActorRolloutRefWorker.wait_on_save_thread()
    • CriticWorker
      • CriticWorker.__init__()
      • CriticWorker.init_model()
      • CriticWorker.compute_values()
      • CriticWorker.update_critic()
      • CriticWorker.save_checkpoint()
      • CriticWorker.load_checkpoint()
      • CriticWorker.clear_optimizer_state()
      • CriticWorker.wait_on_save_thread()
  • trinity.trainer.verl.ray_trainer module

Module contents

Previous Next

© Copyright 2025, Trinity-RFT Team.

Built with Sphinx using a theme provided by Read the Docs.
Other Versions v: v0.1.1
Tags
v0.1.0
v0.1.1
v0.2.0
v0.2.1
Branches
main (latest)