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

Replace categorical variables in mlrMBO with e.g., integer params with mapping. #54

Open
jozik opened this issue Apr 4, 2019 · 7 comments
Assignees

Comments

@jozik
Copy link
Contributor

jozik commented Apr 4, 2019

No description provided.

@jozik jozik changed the title Replace categorical variables with e.g., integer params with mapping. Replace categorical variables in mlrMBO with e.g., integer params with mapping. Apr 4, 2019
@jozik jozik self-assigned this Apr 4, 2019
@rajeeja
Copy link
Collaborator

rajeeja commented Apr 4, 2019

mlr-org/mlrMBO#454

Issue appears due to insufficient init size as stated above, it has been isolated in the above script.

@jozik
Copy link
Contributor Author

jozik commented Apr 4, 2019

It looks like the embedded link in link above (mlr-org/mlrMBO#454)
Doesn't actually link to this file (so I'm including it here):
https://github.com/rajeeja/mlrmbo-bug/blob/master/mlrMBOMixedIntegerTest11a.R

@rajeeja
Copy link
Collaborator

rajeeja commented Apr 12, 2019

I've implemented the solution mentioned, but error persists, it has to do with setting up learner. mlrMBO repo chain:
mlr-org/mlrMBO#454

@jozik
Copy link
Contributor Author

jozik commented Apr 12, 2019

I'm getting the following error when trying to run the example you provided:

Error in isScalarValue(x, na.ok, null.ok, "numeric") :
argument "resolution" is missing, with no default

When I comment out this line:

d1 = generateGridDesign(par.set, trafo = TRUE)

The error is avoided and the code runs without issue.

@rajeeja
Copy link
Collaborator

rajeeja commented Apr 13, 2019

thanks, pushed a fix.

@jozik
Copy link
Contributor Author

jozik commented Apr 13, 2019

Do you still see the error? Or does ensuring that all the categorical variables are present take care of it?

@rajeeja
Copy link
Collaborator

rajeeja commented Apr 13, 2019

Ensuring all categorical variable works only with NULL as learner.
The example is to show the error with surr.rf as learner.

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

2 participants