Simple light-weight flask app for grading homeworks, scoring applications, reviewing papers, etc. A web-server runs locally and starts a browser to allow you to step through documents, showing you a two-pane view with the document on the left and a comment+score box on the right, with easy insertion of lines from comments in other documents (useful for commenting on common mistakes when grading homeworks).
Requires python3 and the flask and markdown packages installed. Easiest way to install is using pip:
$ pip3 install fleval
This will install fleval with the required dependencies.
Alternatively, just clone this repo, and add it to your path, and manually installing the dependencies (pip3 install flask markdown). This is a better option if you want to customize the scripts for your own requirements.
usage: fleval [-h] [-ext EXT] [-nb] jsfile
positional arguments:
jsfile Name of JSON file that comments and scores will be saved in.
optional arguments:
-h, --help show this help message and exit
-ext EXT Extension of files to search for as submissions (e.g., html).
Default pdf.
-nb Do not start browser, or close app when browser window closed.
Note that EXT can be anything your browser can display (pdf, html, txt). Run this in a directory, and fleval will help you step through *.EXT. The basenames of the files will be used as IDs. For example, to grade homework submissions, place all submissions in the same directory as student.id1.pdf, student.id2.pdf, etc.
flerge is a script to help aggregate scores for different questions from different graders.
USAGE: flagg.py out_dir [title]:file.js[,file2.js,...] [title2:file3.js,file4.js ...]
Say you have multiple questions (q1, q2, q3) and TAs (ta1,ta2,ta3), and you ask them to save their scores for different questions for different subsets of students in q?.ta?.js. Moreover, you yourself store general comments (possibly with scores) in general.js and latenote.js. Then, you can run flerge as
$ flerge out/ :general.js,latenote.js 'Question 1':q1.*.js 'Question 2':q2.*.js 'Question 3':q3.*.js > out.csv
This will create a id.html for each student / submission in the directory out/ collating scores from all the different json files, as well as print out a csv file with scores (per question, and total) for every submission.