[Bug]: Machine Learning-the prediction tables crashed

[vladimirsim]

JASP Version

0.19.1

Commit ID

No response

JASP Module

Machine Learning

What analysis are you seeing the problem on?

Prediction

What OS are you seeing the problem on?

Windows 10

Bug Description

The prediction tables did not load.

Expected Behaviour

The prediction tables should load.

Steps to Reproduce

1. Open Machine learning > Prediction
2. Load a trained model based on JASP's Machine learning > Student grades internal data library. The trained model is attached.
3. Load the features from the trained model.
4. The engine crashes right before the last feature has to be loaded.
   ...
   p.zip

Log (if any)

This analysis terminated unexpectedly.

Error in levels<-(*tmp*, value = as.character(levels)): factor level [2] is duplicated

Stack trace
analysis(jaspResults = jaspResults, dataset = dataset, options = options)

.mlPredictionsTable(model, dataset, options, jaspResults, ready, position = 2)

.mlPredictionsState(model, dataset, options, jaspResults, ready)

createJaspState(.mlPredictionGetPredictions(model, dataset))

jaspStateR$new(object = object, dependencies = dependencies)

initialize(...)

.mlPredictionGetPredictions(model, dataset)

.mlPredictionGetPredictions.gbm(model, dataset)

gbm:::predict.gbm(model, newdata = dataset, n.trees = model[['n.trees']], type = 'response')

factor(x[, i], levels = object$var.levels[[i]])

To receive assistance with this problem, please report the message above at: https://jasp-stats.org/bug-reports

More Debug Information

-------- Application Info --------
JASP Version: JASP 0.19.1
Build Branch: lockfiletesting
Build Date: Sep 9 2024 19:00:04 (Netherlands)
Last Commit: 8ac10c9c7aec029fdd695262d7bb4c3742a8ec89

-------- Basic Info --------
Operating System: Windows 10 Version 22H2
Product Version: 10
Kernel Type: winnt
Kernel Version: 10.0.19045
Architecture: x86_64
Install Path: E:/Program Files/JASP
Platfotm Name: windows
System Local: bg_BG

-------- Extra Info --------
Current code page
Active code page: 866
Active code page: 65001

Final Checklist

• I have included a screenshot showcasing the issue, if possible.
• I have included a JASP file (zipped) or data file that causes the crash/bug, if applicable.
• I have accurately described the bug, and steps to reproduce it.

[tomtomme]

@vladimirsim
thx for the report. I can confirm the error with 0.19.1 flatpak.

@koenderks
Can you have a look? The error message appears the moment I add the last variable in the list, called "absences". Its the only metric one. But switching it to ordinal does also not help.

[koenderks]

@tomtomme It should work if you set all ordinal variables to scale like they are in the original dataset (not sure why they are ordinal here?)

[tomtomme]

I loaded the jasp File from the bib. Those are the scales I get. We had a bug in the past where changed scales we're not saved. Maybe this is it. I will investigate.

[tomtomme]

@koenderks
Jasp should indicate that scale is needed, right? But there is No scales Icon in the Box you drop the vars into.

[koenderks]

The error occurs because the model is trained with these variables set to scale, so the mismatch between the trained variables which were set to scale and the current variables which are set to ordinal creates this issue. In the next version, you can only enter scale or nominal variables in this box.