Error at h2o_automl?

[wingraver]

Hello Bernado, your package certainly has great potential use however I cannot it to work. With all attempts I get an error at "_results <- lares::h2o_automl(df = dfm, seed = seed, maxtime = 60)" or any such statement. The error message received = "Error in h2o::h2o.automl(x = setdiff(names(df), "tag"), y = "tag", training_frame = as.h2o(train), : unused argument (exclude_algos = c("StackedEnsemble", "DeepLearning"))". I have already successfully ran the update for the package so I'm wondering if the call to h2o is broken???

[laresbernardo]

Hello @wingraver Thanks for getting in contact! I've just checked h2o's documentation and there shouldn't be a problem with that parameter. The exclude_algos simply excludes some algorithms: http://docs.h2o.ai/h2o/latest-stable/h2o-docs/automl.html Please, try closing and opening RStudio again and retry. If you have just updated de library, you should restart your session (lares or h2o). Let me know if that helped!

[wingraver]

Hi @laresbernardo thanks for the quick response. I'm still getting the error even after closing/opening RStudio, restarting the session multiple times and running the update again. I am able to successfully execute h2o.automl from the h2o package (using a the diabetes dataset as a test) so I'm not sure what's happening between h2o_automl and get the h2o package?

Would my session info help? R 3.4.3 on Windows 7 x64

[laresbernardo]

Hey @wingraver Must be something in your configurations. I have just run this and everything ran as it should. Can you try it?

require(dplyr)
data(Diabetes, package="heplots")
df <- Diabetes %>% rename(tag = group)
results <- lares::h2o_automl(df, max_time = 30)

[wingraver]

Stranger and stranger... I reloaded my environment and tried installing the "heplots" package... I get an error message "unable to access index for repository http://www.stats.ox.ac.uk/pub/RWin/src/contrib: Line starting ' ...' is malformed!"? Could it an issue with the version of RStudio (currently on 1.1.456)

[laresbernardo]

From here it seems that way @wingraver I currently have the 1.1.463 version and R version 3.5.1 Btw, I used that package/dataset because is the diabetes' dataset I found.. isn't it the one you are using?

[wingraver]

Okay so maybe it's got to do with the package versions... I'm using R version 3.4.3. The diabetes dataset was one you had mentioned in an article PimaIndianDiabetes (from the mlbench). Here's my session info...

R version 3.4.3 (2017-11-30) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows 7 x64 (build 7601) Service Pack 1

Matrix products: default

locale: [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252 [4] LC_NUMERIC=C LC_TIME=English_United States.1252

attached base packages: [1] stats graphics grDevices utils datasets methods base

other attached packages: [1] lares_4.3 dplyr_0.7.4 RevoUtilsMath_10.0.1 RevoUtils_10.0.7 RevoMods_11.0.0 MicrosoftML_9.3.0
[7] mrsdeploy_1.1.3 RevoScaleR_9.3.0 lattice_0.20-35 rpart_4.1-11

[laresbernardo]

What happens when you run lares::updateLares() ?

[wingraver]

It ran successfully (just now). I then restart RStudio but when I get to the lares::h2o_automl statement I still get the error.

[laresbernardo]

Try installing this as well: devtools::install_github("r-lib/rlang", build_vignettes = TRUE)

[wingraver]

Hi @laresbernardo thanks for all the additional feedback so far... still no improvement. Maybe I should have the IT team upgrade to R 3.5.1?

[laresbernardo]

Yes, I'd strongly recommend that. Let me know if that gets it fixed! I'll keep this ticket open util we finally understand what happened ;)