R Notebook Format¶
Otter Assign is compatible with Otter’s R autograding system and currently supports Jupyter notebook master documents. The format for using Otter Assign with R is very similar to the Python format with a few important differences.
Assignment Metadata¶
As with Python, Otter Assign for R also allows you to specify various assignment generation arguments in an assignment metadata cell.
```
BEGIN ASSIGNMENT
init_cell: false
export_cell: true
...
```
This cell is removed from both output notebooks. Any unspecified keys will keep their default values. For more information about many of these arguments, see Usage and Output. The YAML block below lists all configurations supported with R and their defaults. Any keys that appear in the Python section but not below will be ignored when using Otter Assign with R.
requirements: requirements.txt # path to a requirements file for Gradescope; appended by default
overwrite_requirements: false # whether to overwrite Otter's default requirements rather than appending
environment: environment.yml # path to custom conda environment file
template_pdf: false # whether to generate a manual question template PDF for Gradescope
generate: # configurations for running Otter Generate; defaults to false
points: null # number of points to scale assignment to on Gradescope
threshold: null # a pass/fail threshold for the assignment on Gradescope
show_stdout: false # whether to show grading stdout to students once grades are published
show_hidden: false # whether to show hidden test results to students once grades are published
files: [] # a list of file paths to include in the distribution directories
Autograded Questions¶
Here is an example question in an Otter Assign for R formatted notebook:
For code questions, a question is a description Markdown cell, followed by a solution code cell
and zero or more test code cells. The description cell must contain a code block (enclosed in
triple backticks ```) that begins with BEGIN QUESTION on its own line, followed by YAML that
defines metadata associated with the question.
The rest of the code block within the description cell must be YAML-formatted with the following fields (in any order):
name(required) - a string identifier that is a legal file name (without an extension)manual(optional) - a boolean (defaultfalse); whether to include the response cell in a PDF for manual gradingpoints(optional) - a number or list of numbers for the point values of each question. If a list of values, each case gets its corresponding value. If a single value, the number is divided by the number of cases so that a question with \(n\) cases has test cases worth \(\frac{\text{points}}{n}\) points.
As an example, the question metadata below indicates an autograded question q1 with 3 subparts
worth 1, 2, and 1 points, respectively.
```
BEGIN QUESTION
name: q1
points:
- 1
- 2
- 1
```
Solution Removal¶
Solution cells contain code formatted in such a way that the assign parser replaces lines or portions of lines with prespecified prompts. The format for solution cells in R notebooks is the same as in Python notebooks, described here. Otter Assign’s solution removal for prompts is compatible with normal strings in R, including assigning these to a dummy variable so that there is no undesired output below the cell:
# this is OK:
. = " # BEGIN PROMPT
some.var <- ...
" # END PROMPT
Test Cells¶
The test cells are any code cells following the solution cell that begin with the comment
## Test ## or ## Hidden Test ## (case insensitive). A Test is distributed to students
so that they can validate their work. A Hidden Test is not distributed to students, but is used
for scoring their work. When writing tests, each test cell maps to a single test case and should
raise an error if the test fails. The removal behavior regarding questions with no solution
provided holds for R notebooks.
## Test ##
testthat::expect_true(some_bool)
## Hidden Test ##
testthat::expect_equal(some_value, 1.04)