I'm trying to test moodle's machine learning model for our research. I used the student dropout as a reference. When using php backend I get a feedback regarding the accuracy, however when I try to switch to python backend it returns an error.
Error:
Debug info:
Error code: errornopredictresults
×Stack trace:
line 568 of /lib/mlbackend/python/classes/processor.php: moodle_exception thrown
line 314 of /lib/mlbackend/python/classes/processor.php: call to mlbackend_python\processor->exec_command()
line 654 of /analytics/classes/model.php: call to mlbackend_python\processor->evaluate_classification()
line 217 of /admin/tool/analytics/model.php: call to core_analytics\model->evaluate()
Here's the temp data being fed to the python backend:
timesplitting,nfeatures,moodleversion,targetcolumn,targettype,targetclasses
"\core\analytics\time_splitting\no_splitting",3,2021051701.02,"\local_learningstyle\analytics\target\active_reflective",discrete,"[0,1]"
"\local_learningstyle\analytics\indicator\active_reflective\chats","\local_learningstyle\analytics\indicator\active_reflective\discussions","\local_learningstyle\analytics\indicator\active_reflective\exercises","\local_learningstyle\analytics\target\active_reflective"
-1,-1,-1,0
1,1,-1,1
-1,-1,-1,1
1,-1,-1,1
-1,1,-1,0
Can someone shed some light on this?
Moodle 3.11.1+ (Build: 20210723)
moodle-mlbackend==2.4.0 (moodle says it can't install latest)