ktdg 0.1.13

Library to simulate knowledge tracing datasets

ktdg (Knowledge tracing data generator)

Library used to create synthetic knowledge tracing data. Example configs can be found in config.

Usage | Setup | Documentation

Usage

To create a new config or complete an existing one:

$ ktdg create --help
Usage: ktdg create [OPTIONS] CONFIG

  (c) Creates a config or completes it, saving it to the given file.

Arguments:
  CONFIG  Path of the config to complete or create  [required]

Options:
  -h, --help  Show this message and exit.

To generate the synthetic data from the config:

$ ktdg generate --help
Usage: ktdg generate [OPTIONS] CONFIG

  (g) Generates the data for the given config, saving it as a json file named
  "data.json".

Arguments:
  CONFIG  Configuration file to use  [required]

Options:
  -h, --help  Show this message and exit.

Setup

1. Install poetry

2. poetry config virtualenvs.in-project true

3. poetry install

4. source .venv/bin/activate

Documentation

Distributions

constant: All samples have the same value value.

normal: Samples are taken from a normal distribution with mean mu and standard deviation sigma.

binomial: Samples are taken from a binomial distribution with number of possible successes n and probability of success p.