alangpierce

#!/usr/bin/env python
import argparse
from collections import Counter, namedtuple
import json
import os
import subprocess
import sys
import textwrap
import urllib
import webbrowser

alangpierce

package org.khanacademy.android.graphgen;

import org.khanacademy.android.dependencies.components.ApplicationComponent;

import dagger.Component;
import dagger.Provides;

import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.InputStream;

alangpierce

Snapshot of Khan Academy's BigBingo A/B testing framework and related code.

Here's a basic overview:
-summarize.py is the most interesting file. It contains all stages of the
summarize task, as well as the publish and archive steps that happen at the
end.
-bq_pipelines.py contains lots of useful pipelines for interacting with
BigQuery. QueryToTableBatchPipeline can run many simultaneous queries, and will
properly handle all batching and retry logic.
-config.py is where all experiment configuraiton lives. For this Gist, I

alangpierce

#!/bin/sh

while [ 1 ]
do
    ./refresh_bigquery_token.sh
    sleep 10
done