search

XanderHorn / autoML

Automated machine learning in R
MIT License
9 stars 7 forks source link

error using example in AutoLearn #4

Closed theiman112860 closed 4 years ago

theiman112860 commented 5 years ago

Hi Xander,

Very cool project!! I ran the following example using autoLearn:

mod <- autoLearn(train = iris, target = "Species")

and get the following errors:

Error in stopWithJobErrorMessages(inds, vcapply(result.list[inds], as.character)) : Errors occurred in 1 slave jobs, displaying at most 10 of them:

00001: Error in lognet(x, is.sparse, ix, jx, y, weights, offset, alpha, nobs, : one multinomial or binomial class has 1 or 0 observations; not allowed

Any ideas on what to do? Thank you!

XanderHorn commented 5 years ago

I see the problem comes in that the iris data set is too small for the lasso, ridge and mixture model to find the optimal lambda value using cross validation and thus there are not enough outcomes in the folds.

I would suggest to replicate the iris dataset a few times and append, then run the example again with possible succession.

Please also keep a close eye on the following repo as it will replace autoML and autoEDA and resolve some of the model fitting issues: https://github.com/XanderHorn/lazy

Please let me know if I can close this issue?

theiman112860 commented 5 years ago

Hi Xander, Thank you!!  Really awesome project!! Sincerely, tom

-----Original Message----- From: Xander Horn notifications@github.com To: XanderHorn/autoML autoML@noreply.github.com Cc: theiman112860 theiman@verizon.net; Author author@noreply.github.com Sent: Fri, Jul 26, 2019 12:25 pm Subject: Re: [XanderHorn/autoML] error using example in AutoLearn (#4)

I see the problem comes in that the iris data set is too small for the lasso, ridge and mixture model to find the optimal lambda value using cross validation and thus there are not enough outcomes in the folds.I would suggest to replicate the iris dataset a few times and append, then run the example again with possible succession.Please also keep a close eye on the following repo as it will replace autoML and autoEDA and resolve some of the model fitting issues: https://github.com/XanderHorn/lazy— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

theiman112860 commented 4 years ago

Hi Xander, Is your Lazy package still available? I did not see it on GitHub.. Sincerely, tom

-----Original Message----- From: Xander Horn notifications@github.com To: XanderHorn/autoML autoML@noreply.github.com Cc: theiman112860 theiman@verizon.net; Author author@noreply.github.com Sent: Tue, Nov 12, 2019 4:22 pm Subject: Re: [XanderHorn/autoML] error using example in AutoLearn (#4)

Closed #4.— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.