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

Make stochastic functions stochastically deterministic #3

Open
sdvillal opened this issue Oct 24, 2014 · 2 comments
Open

Make stochastic functions stochastically deterministic #3

sdvillal opened this issue Oct 24, 2014 · 2 comments
Assignees
Labels

Comments

@sdvillal
Copy link
Collaborator

At the moment hctsa uses the system rng without controlling seeds, thus making impossible to control the output of stochastic operators. Ben would take care of this at some point in matlab land.

In in python land we will need to generate the bindings to stochastic operators to accept a rng seed and write a common unit test to catch non-determinism. @benfulcher can I assume that all the stochastic features are tagged as "stochastic"?

@sdvillal sdvillal added the bug label Oct 24, 2014
@sdvillal sdvillal self-assigned this Oct 24, 2014
@benfulcher
Copy link

Yeah I ran a test over all operations (running each I think 50 times) for a handful of test time series to produce a basic labeling of all operations as "stochastic" or not. It's not an easy process, because sometimes the stochasticity depends on the length of a time series (e.g., long timeseries are subsampled randomly), and sometimes the randomness comes at the level of rounding errors, or could potentially depend on the platform or version of Matlab. I will try to resolve this issue "in Matlab land" by the year's end.

@sdvillal
Copy link
Collaborator Author

Work towards working out this in this HCTSA issue

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

No branches or pull requests

2 participants