Pepijn Kooijmans 1f7e7b4a90 fix: close envs between tasks to prevent worker process accumulation ...

eval_policy_all never closed environments after each task completed,
causing AsyncVectorEnv worker processes to accumulate (N_tasks × n_envs).
This led to OOM, BrokenPipeError and EOFError on multi-task benchmarks.

Also fixes:
- AsyncVectorEnv compat in envs/utils.py (use get_attr/call instead of .envs)
- Tuple task handling in tokenizer_processor and lerobot_eval
- _LazyAsyncVectorEnv for deferred worker spawning in LIBERO

Made-with: Cursor

2026-04-07 12:30:22 +02:00

..

test_dispatch.py

fix: close envs between tasks to prevent worker process accumulation

2026-04-07 12:30:22 +02:00

test_envs.py

feat(sim): EnvHub - allow loading envs from the hub (#2121)

2025-11-04 14:52:46 +01:00