Wrappers and support for bsuite

描述

貢獻者指南

技術棧

python

領域

machine learning

議題類型

feature

難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。

3

預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。

3-5 days

活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。

stale

清晰度議題是否清楚說明預期改動、驗收標準和下一步。

needs investigation

前置要求

familiarity with garageunderstanding of bsuite

新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。

10

研究方向

The issue links to bsuite, a benchmark suite for RL. To add support, one should first understand how garage currently handles benchmarks and environments. Look at existing experiment configurations and environment wrappers in the garage codebase. The implementation will likely involve creating wrappers that conform to garage's environment interface and setting up experiment scripts to run bsuite tasks. There are no comments or PRs to guide the work, so the contributor must design the integration independently.