OpenRL-Lab / openrl

Unified Reinforcement Learning Framework
https://openrl-docs.readthedocs.io
Apache License 2.0
640 stars 62 forks source link

deepspeed support #282

Closed WentseChen closed 10 months ago

WentseChen commented 10 months ago

Description

Types of changes

Checklist

codecov[bot] commented 10 months ago

Codecov Report

Attention: 216 lines in your changes are missing coverage. Please review.

Comparison is base (38da73a) 74.49% compared to head (fc02030) 73.38%. Report is 3 commits behind head on main.

Files Patch % Lines
openrl/modules/networks/policy_network_gpt.py 0.00% 112 Missing :warning:
openrl/modules/networks/value_network_gpt.py 0.00% 72 Missing :warning:
openrl/envs/nlp/rewards/intent.py 50.00% 10 Missing :warning:
openrl/envs/nlp/rewards/kl_penalty.py 68.96% 9 Missing :warning:
openrl/utils/logger.py 18.18% 9 Missing :warning:
openrl/algorithms/ppo.py 55.55% 4 Missing :warning:
Additional details and impacted files ```diff @@ Coverage Diff @@ ## main #282 +/- ## ========================================== - Coverage 74.49% 73.38% -1.12% ========================================== Files 252 254 +2 Lines 13035 13256 +221 ========================================== + Hits 9711 9728 +17 - Misses 3324 3528 +204 ``` | [Flag](https://app.codecov.io/gh/OpenRL-Lab/openrl/pull/282/flags?src=pr&el=flags&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=OpenRL-Lab) | Coverage Δ | | |---|---|---| | [unittests](https://app.codecov.io/gh/OpenRL-Lab/openrl/pull/282/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=OpenRL-Lab) | `73.38% <16.92%> (-1.12%)` | :arrow_down: | Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=OpenRL-Lab#carryforward-flags-in-the-pull-request-comment) to find out more.

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.