Issues

• r

  Emerge-Lab/gpudrive#537 · Filter by author stafimleandro-create opened on May 1, 2026

• The problem of running out of the lane

  Emerge-Lab/gpudrive#533 · Filter by author wangbingke0 opened on Jan 23, 2026

• 00_simple_model_test.ipynb how to use sb3 model zip?

  Emerge-Lab/gpudrive#531 · Filter by author Tobias-lt opened on Dec 30, 2025
  1

• Issue with Importing madrona_gpudrive in Docker Setup

  Emerge-Lab/gpudrive#529 · Filter by author niuyuchenf opened on Nov 30, 2025
  2

• get_expert_actions() returns all-zero actions when using Waymo dataset — imitation learning impossible

  Emerge-Lab/gpudrive#528 · Filter by author duguxiaohun opened on Nov 18, 2025

• evaluating the policy in single agent mode

  Emerge-Lab/gpudrive#527 · Filter by author MohammadSajadMarvi opened on Nov 10, 2025

• viewer is displaying abnormally with 'lingering' shadows from expert-type agents.

  Emerge-Lab/gpudrive#522 · Filter by author XiangyueGao opened on Sep 20, 2025

• Clarification on segment length and width in map observations

  Emerge-Lab/gpudrive#521 · Filter by author itsbman opened on Sep 18, 2025

• Compatibility with 3090/4090 Series

  Emerge-Lab/gpudrive#519 · Filter by author alfredgu001324 opened on Sep 9, 2025
  2

• Traffic light support

  Emerge-Lab/gpudrive#518 · Filter by author daphne-cornelisse opened on Sep 8, 2025

• Long-tail scenario manual setup

  Emerge-Lab/gpudrive#516 · Filter by author ChichengZZZ opened on Aug 25, 2025
  1

• No module named 'pufferlib.cleanrl'

  Emerge-Lab/gpudrive#514 · Filter by author wangbingke0 opened on Aug 17, 2025
  3

• NVRTC_ERROR_COMPILATION

  Emerge-Lab/gpudrive#513 · Filter by author oorleck opened on Aug 11, 2025

• Time step length mismatch

  Emerge-Lab/gpudrive#512 · Filter by author wangbingke0 opened on Aug 11, 2025
  2

• Baseline PPO Pufferlib script passes None as seed to SceneDataLoader

  Emerge-Lab/gpudrive#510 · Filter by author cevahir-koprulu opened on Jul 30, 2025

• CUDA_ERROR_LAUNCH_OUT_OF_RESOURCES during ppo_sb3.py execution

  Emerge-Lab/gpudrive#509 · Filter by author Ibrahim-Sakaama opened on Jul 27, 2025
  2

• Question: How to reproduce paper results and current best configuration

  Emerge-Lab/gpudrive#508 · Filter by author rshimomura0815 opened on Jul 25, 2025
  9

• how to use trained model

  question
  Emerge-Lab/gpudrive#507 · Filter by author wangbingke0 opened on Jul 23, 2025
  1

• done flag is delayed by one step after collision (reward = –0.75)

  Emerge-Lab/gpudrive#504 · Filter by author KexianShen opened on Jul 22, 2025
  1

• Add support for hparam tuners

  Emerge-Lab/gpudrive#484 · Filter by author eugenevinitsky opened on Jun 26, 2025

• Add stochasticity to dynamics and observations

  Emerge-Lab/gpudrive#483 · Filter by author eugenevinitsky opened on Jun 26, 2025

• Remove cost of resets when resetting to same world

  Emerge-Lab/gpudrive#482 · Filter by author eugenevinitsky opened on Jun 26, 2025

• Add support for procedural agent position generation

  Emerge-Lab/gpudrive#481 · Filter by author eugenevinitsky opened on Jun 26, 2025

• Add multi-GPU support

  Emerge-Lab/gpudrive#480 · Filter by author eugenevinitsky opened on Jun 26, 2025

• Merge jerk model

  Emerge-Lab/gpudrive#479 · Filter by author eugenevinitsky opened on Jun 26, 2025