Skip to content
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

Precompilation prohibitively long #237

Closed
AlexLewandowski opened this issue Apr 14, 2021 · 2 comments · Fixed by JuliaReinforcementLearning/ReinforcementLearningEnvironments.jl#139

Comments

@AlexLewandowski
Copy link
Contributor

Precompilation for ReinforcementLearningEnvironments is long - longer than other, more complex, packages like Flux. I'm not sure if this is due to the sparsely used dependencies, like OrdinaryDiffEqs and GR, or maybe due to the @require statements for the optional envs. In any case, I think some of these dependencies can be removed. The ODE package is used only once to solve runge-kutta equations in acrobot and GR is used to plot mountaincar and cartpole. While these featuers are useful, I am not sure if it warrants such long precompilation times.

@findmyway findmyway transferred this issue from JuliaReinforcementLearning/ReinforcementLearningEnvironments.jl Apr 15, 2021
@findmyway
Copy link
Member

Thanks for pointing this out. This has been a headache to me for a long time and I was intended to tackle it after [email protected]
Let me investigate all the dependencies first.
Personally, I prefer to move some parts of them (like ODE and GR) into lazy loading.

@findmyway
Copy link
Member

The precompile time is reduced to ~2s now.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants