AashVed/DedeuceRL: DedeuceRL v1.0.2
Abstract
Added dedeucerl-eval-parallel to run shard-parallel evals and merge results into a single JSONL. Added episode selection + sharding to dedeucerl-eval via --episodes and --shard. Added split-aware resume support via --resume (and --append). Added split hashing and episode utilities (parse_index_spec, parse_shard, apply_shard, compute_split_hash). Added tests ensuring parallel output matches sequential output.
Fields of Study & Subjects
Reinforcement Learning in Robotics
Control Systems and Identification
Machine Learning and Algorithms
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 17 January 2026