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

Still an infinite loop in bisect #11

Open
marius311 opened this issue Feb 12, 2021 · 8 comments
Open

Still an infinite loop in bisect #11

marius311 opened this issue Feb 12, 2021 · 8 comments

Comments

@marius311
Copy link
Contributor

Sorry to sound like a broken record and I know we discussed in #10 but there's still an infinite loop path in bisect. I confirm because I run into it sporadically, and editing this line to read if c.dϕ >= 0 || numfg>10 to break out if it tries more than 10 bisect steps fixes it for me.

No doubt I have some numerically noisy gradients causing this, and I'm happy if the package sends me a warning or even errors when bisect fails due to that, but just infinite looping really kills my analysis (eg I run 100 sims and step away from my computer for a while, only to come back and find sim 3 got stuck in an infinite loop and it never went further).

@Jutho
Copy link
Owner

Jutho commented Feb 12, 2021

I am happy to accept a PR, but I would just make a separate branch in the if for the case numfg>10 and print out a warning indeed.

@ArtemStrashko
Copy link

Hi, it seems I have exactly the same issue and in my case it would be useful if the optimizer returned something (e.g., just an initial guess x_0). Is there some simple way to force it to stop bisecting after a few attempts and give up (just returning an initial condition)?
Thank you.

@Jutho
Copy link
Owner

Jutho commented Feb 10, 2022

I will try to look at this soon; the linesearch logic should be changed somewhat to allow for a graceful return indeed.

@maartenvd
Copy link
Contributor

out of curiosity, is there a small minimal example somewhere of bisect looping?

@ArtemStrashko
Copy link

Oh, I am afraid not really right now. I am developing an MPS-like architecture for image classification and when I optimize MPS tensors one by one, sometimes optimizations just gets stuck on some tensor and does not proceed to the next tensors.

@ArtemStrashko
Copy link

Hi, if I just download OptimKit.jl and then add this condition numfg>10 as suggested by @marius311, could you please suggest how I then can use it? If I just include a line include("../../OptimKit/src/OptimKit.jl"), I then get an error UndefVarError: ConjugateGradient not defined. But if I also add another line include("../../OptimKit/src/gd.jl"), I get another error LoadError: UndefVarError: AbstractLineSearch not defined. Finally, if I include the following lines

using Printf
include("../../OptimKit/src/OptimKit.jl")
include("../../OptimKit/src/linesearches.jl")
include("../../OptimKit/src/gd.jl")

I get this error: LoadError: UndefVarError: OptimizationAlgorithm not defined (and I always get WARNING: replacing module OptimKit, even though I ran Pkg.rm("OptimKit.jl) before).
Thank you.

@marius311
Copy link
Contributor Author

One hack is just to do

using OptimKit
@eval OptimKit function bisect(iter::HagerZhangLineSearchIterator, a::LineSearchPoint, b::LineSearchPoint)
  # modified body here...
end

Another is you could pkg> dev --local OptimKit and edit the files directly.

@ArtemStrashko
Copy link

Thank you @marius311! I use the first approach (with @eval). Sorry for my ignorance, but how should I proceed with the second approach? I mean, I execute pkg> dev --local OptimKit, but what's next? One part of an output of the previous command is OptimKit v0.3.1 ⇒ v0.3.2 dev/OptimKit, but when I cd to \dev, I cannot find OptimKit there.
Thank you!

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

No branches or pull requests

4 participants