wlmiller / rmlessontiming

Code for estimating lesson length using features extracted from lesson scripts.

RM Lesson Timing

The main code for calculating lesson time is lessontiming.py.

This code estimates lesson lengths for RM lessons. The code asks the user to select the OS file for the lesson; it requires the OS file name to include the lesson number and the lesson item scripts to be located in a directory called Scripts/ in the same directory as the OS file. The OS file and all lesson item scripts must be in the docx format.

Python 2.7 is required. There is one optional command line argument; the path to the OS file can be supplied directly, skipping the file selection dialog.

The included Python scripts are:

1. lessontiming.py: Given an OS file, extracts the "weak + behind" and "weak + ontime" paths and estimates the length of each lesson item as well as the total length of these two paths. The resulting data are placed in file called [lesson]_timing.csv in the same directory as the OS file, and this file is automatically opened for the user.
2. parseOSfile.py: Called by lessontiming.py and getallpaths.py, extracts the lesson paths from a given OS file.
3. lessonitemstats.py: Called by lessontiming.py and parseallitems.py, extracts the relevant features in the given lesson item script.
4. getallpaths.py: Loops through all lessons in the curriculum and extracts the "weak + behind" path using parseOSfile.py.
5. parseallitems.py: Loops through all lesson items in the curriculum and extracts the relevant statistics using lessonitemstats.py.

Note: Some of the feature extraction methods may seem somewhat tortured. This is because the format of OS files and scripts is almost, but not quite, standard.