WQET Grader Client

Installation

pip install wqet-grader

Usage: Notebooks

Simple grading:

import wqet_grader

submission = "My Answer"
wqet_grader.grade('Sample Assessment', 'Question 1', submission)

Object grading:

import wqet_grader

arg1 = 3.14
arg2 = "something else"
wqet_grader.grade_object('Sample Assessment', 'Question 1', arg1=arg1, arg2=arg2)

Supported objects:

file Binary file
pandas_dataframe Pandas DataFrame object
sklearn_model SKLearn pipeline/model
float, int, string, dict, list - Primative types

Usage: Content Development Debugging

You can load a directory of grading algorithm modules, by running the following command:

./scripts/load_and_run fixtures/content

This will automatically restart the server when any changes to the Grading API code or content is detected. Reporting to the WQET platform is disabled and additional debug logging is enabled on the console.

The sample notebook of this content can be run as follows:

cd fixtures/notebooks
./run.sh

Then open: http://127.0.0.1:8888/lab/tree/sample-questions.ipynb?token=test

Usage: Content Integration Testing

You can spin up a grading server for a directory of content, using the following code:

from wqet_grader.utils import set_grading_content_path
set_grading_content_path("my-curriculum-folder")

from wqet_grader.server import app
app.run(host="localhost", port=2400, debug=True, threaded=False)

Configuration

Through environment variables:

GRADING_API_URL - URL of grading API
VM_TOKEN - Temporary token for user authentication
SCORE_OUTPUT_FORMAT - Format of score output in notebook: json or html (defaults to JSON)

Development Setup

Requirements:

Python 3.6+.

Setup your Virtual Environment:

make venv

Testing

To run unit tests:

make test

Packaging

make package.build
make package.release

dominiek/wqet-grader