mirror of
https://github.com/NousResearch/atropos.git
synced 2026-04-28 17:29:30 +00:00
Properly exported SHM primitives in atroposlib.api. Added technical documentation for the SkyRL bridge and a dedicated local test runner (test_shm.sh). Verified with library tests.
1.2 KiB
1.2 KiB
SkyRL Integration (SHM Transport)
This directory contains the skyrl_adapter.py, which enables Atropos to act as a high-performance reasoning environment provider for the SkyRL training framework.
Architecture
The integration utilizes a Zero-Copy Shared Memory (SHM) transport to eliminate the "JSON Tax" during reasoning-dense RL collection.
- Transport:
atroposlib.api.shm_buffer.ZeroCopySHMBuffer - Adapter:
atroposlib.envs.skyrl_adapter.SkyRLAdapter
Performance
Benchmarks on RTX 3090 hardware show an ~8x throughput gain compared to standard HTTP/JSON transport:
- Baseline (HTTP): ~2,000 trajectories/sec
- Hardened (SHM): 16,500+ trajectories/sec
Usage
To enable the SHM transport, initialize the environment with TransportType.SHM:
from atroposlib.envs.base import TransportType
from atroposlib.envs.skyrl_adapter import SkyRLAdapter
env = SkyRLAdapter(
transport=TransportType.SHM,
shm_name="atropos_shm_run1",
# ... other config
)
Testing
A dedicated end-to-end verification script for the SHM bridge is available in the root directory:
bash test_shm.sh
This script verifies the atomic index synchronization and data integrity without requiring a full GPU cluster.