error on plot predict_parts duplicate row names

[jamesburnsTTRS]

explainer_2 <- explain_xgboost(clf, data = newmat, y=Weekly_SL_Agg$LN_SWP_Act)
Preparation of a new explainer is initiated
-> model label : xgb.Booster ( default )
-> data : 378 rows 49 cols
-> target variable : 378 values
-> predict function : yhat.xgb.Booster will be used ( default )
-> predicted values : No value for predict function target column. ( default )
-> model_info : package xgboost , ver. 1.7.6.1 , task regression ( default )
-> predicted values : numerical, min = zz.373191 , mean = kk.75155 , max = aa.02619
-> residual function : difference between y and yhat ( default )
-> residuals : numerical, min = -z.7959752 , mean = -b.000319823 , max = f.2809396
The model works
plot(predict_parts(explainer_2,newmat))
Error in .rowNamesDF<-(x, value = value) :
duplicate 'row.names' are not allowed
In addition: Warning message:
non-unique value when setting 'row.names': ‘xgb.Booster’

first variable is a predictor 1 week of Black Friday 0 elsewhere for 378 rows of weekly data
and I think that's what's driving this error or do you need to create x axis per variable

there are many variables with dupes per month since many are monthly figures. Do you need to create a separate x axis per variable vs a row id common for all?

[maksymiuks]

Hi,

Could you please share some examples of dummy data an your model so we can reproduce the issue?

[jamesburnsTTRS]

Thanks for the quick response! I think I figured it out. All the parameters were zero and the forecasts were the intercept. I looked at the file created by predict_parts(explainer, data). How did you build that file and what are the various sections?Sent from my iPhoneOn Apr 6, 2024, at 4:32 PM, maksymiuks ***@***.***> wrote: Hi, could you please share some dummy data example so we can reproduce the issue? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: ***@***.***>