search

assume-framework / assume

ASSUME - Agent-based Simulation for Studying and Understanding Market Evolution
https://assume.readthedocs.io
22 stars 5 forks source link

Fixed RL evaluations #280

Closed kim-mskw closed 7 months ago

kim-mskw commented 7 months ago

Fixed RL evaluations to use compare_and_save_policies function instead of comparing in

codecov[bot] commented 7 months ago

Codecov Report

Attention: 15 lines in your changes are missing coverage. Please review.

Comparison is base (8290bf7) 82.59% compared to head (a56c84c) 83.03%.

Files Patch % Lines
assume/reinforcement_learning/learning_role.py 75.00% 7 Missing :warning:
assume/scenario/loader_csv.py 71.42% 4 Missing :warning:
assume/strategies/learning_advanced_orders.py 0.00% 2 Missing :warning:
assume/strategies/learning_strategies.py 80.00% 1 Missing :warning:
assume/world.py 50.00% 1 Missing :warning:
Additional details and impacted files ```diff @@ Coverage Diff @@ ## main #280 +/- ## ========================================== + Coverage 82.59% 83.03% +0.44% ========================================== Files 40 40 Lines 4550 4539 -11 ========================================== + Hits 3758 3769 +11 + Misses 792 770 -22 ``` | [Flag](https://app.codecov.io/gh/assume-framework/assume/pull/280/flags?src=pr&el=flags&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=assume-framework) | Coverage Δ | | |---|---|---| | [pytest](https://app.codecov.io/gh/assume-framework/assume/pull/280/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=assume-framework) | `83.03% <71.15%> (+0.44%)` | :arrow_up: | 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=assume-framework#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.

kim-mskw commented 7 months ago

get it, so are we now saving when each individual reward is max or the average among all agents? what did we decided from out last long discussion on this topic?

we still did it with the average among all agents since we did not decide on an alternative. With Florians upcoming commit it is programmed in such a way, that we can easily change that in the future though.