AUC step generates NA warning - significant?

[shlid007]

I am running the AUC step that calculates the parsimony plot. The log shows this warning and other similar ones - significant? I assume it is not a show-stopper since the number of NAs is relatively small compared to my dataset size.

Warning in compute_auc_val(train_set_1, validation_set_1, variable_list, :
NA in the validation_set_2: 52

[yedimi]

I have a similar error this is error sentence.

Warning messages:
1: In assign_score(validation_set_2, score_table) :
NAs introduced by coercion
2: In print_roc_performance(y_validation, validation_set_3$total_score, :
NA in the score: 1148

[fengx13]

May I check what the sample size of your data is? Does it contain missing values? Is it an error or a warning? This is to say NA was introduced during dataset transformation. You can narrow it down to specific variables causing the problem. (e.g., delete one variable at a time)

[shlid007]

Hi - sorry for late response; I am going to do more data preparation steps to see if that helps the error. I will let you know what I find! In meantime, I'm trying to understand the main advantages/difference with AutoScore compared to other interpretable models. I realize parsimony is a defining attribute, but what about AutoScore's methodology makes it a parsimonious, optimal model? Is it how the random forest variable importance is baked in and variables iteratively selected based on AUC score? Thanks for any feedback you can provide! Sincerely, Sue

…

________________________________ From: XIE FENG ***@***.***> Sent: Friday, June 23, 2023 9:26 PM To: nliulab/AutoScore ***@***.***> Cc: Sue Lhymn ***@***.***>; Author ***@***.***> Subject: Re: [nliulab/AutoScore] AUC step generates NA warning - significant? (Issue #7) May I check what the sample size of your data is? Does it contain missing values? Is it an error or a warning? This is to say NA was introduced during dataset transformation. You can narrow it down to specific variables causing the problem. (e.g., delete one variable at a time) — Reply to this email directly, view it on GitHub<#7 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/A3MFT2NWXVUOOOXC5KWLLLTXMY63HANCNFSM6AAAAAAYLBQNPA>. You are receiving this because you authored the thread.Message ID: ***@***.***>