-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
New experiments features (model/agent/representation comparison, ...) #37
base: master
Are you sure you want to change the base?
Conversation
…esearch/SeaPearlZoo into louis/experiments_fixes
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think we should remove most of the old files to run the experiments (agents*.jl
, features.jl
etc...) and leave only one as an example.
learning_cp/nqueens/rewards.jl
Outdated
@@ -50,5 +50,28 @@ end | |||
|
|||
function SeaPearl.set_reward!(::Type{SeaPearl.EndingPhase}, lh::SeaPearl.LearnedHeuristic{SR,NQueenReward,O}, model::SeaPearl.CPModel, symbol::Union{Nothing,Symbol}) where {SR <: SeaPearl.AbstractStateRepresentation,O <: SeaPearl.ActionOutput} | |||
lh.current_reward += 100 / model.statistics.numberOfNodes | |||
# Getting the number of conflicts on nqueens(20) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It seems to me that the work related to NQueenReward
has been abandoned, not giving good results. The good ideas have been incorporated in the GeneralReward
.
@@ -128,16 +127,16 @@ function trytrain(nbEpisodes::Int) | |||
experienceTime = now() | |||
dir = mkdir(string("exp_",Base.replace("$(round(experienceTime, Dates.Second(3)))",":"=>"-"))) | |||
expParameters = Dict( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The use of ExpParameters has been dropped.
…o into heterogeneous
…o into heterogeneous
What's the status of this PR? Anything interesting we can merge? |
The PR can be merge, it contains many new features for the different experiments we run for the paper. |
No description provided.