engelsmann / boxplot Goto Github PK

When two or more students work and submit together, both their identical scores will appear in the statistics underlying the box plot (quartile set).

• Is this the statistics we wish to present?

Discussed in
my S-O question "How should Python virtual environment be handled on Azure?"
in order to involve external expertise.

Involves i.a. the assignment:

https://www.github.com/engelsmann/boxplot/tree/test_suite_1/templates%2Fboxplot%2Findex.html#L26

The attempts to "raise KeyError", e.g. in

In case they work for first score, validators should be extended to all 6+6 scores

Boxplot regarding 1 class and 1 assignment will be needed as background for as many compound plots (fit overlay) as the class has students. If caching considerably outperforms production, caching should be contemplated.

Is there any performance arguments that only marginal sound of the 6-by-6 scores matrix are stored in AssessmentScores?

• Could e.g. a 6x6 heatmap of the scores help in the feedback dialogue?

Alle tests i <box_whiskers_demo/pseudocode/step1.md> skal implementeres eller fjernes fra step1, før der kan merges.